Then, Elon Musk chimes in to the original whistleblower’s post saying that his posts sound like a set-up. Journalists rightfully cast doubt on how much this whistleblower actually knows, given he was only at Anthropic for a few months. We then remember the Hugging Face incident, and we wonder why regulators would allow any company to IPO if the consensus is that its product would precipitate an apocalypse.

A few days later, Anthropic CEO Dario Amodei publishes a 3,800-word manifesto calling for regulation and restraint in the AI space—something a private company only does if it has some ulterior motive. This manifesto garners public support from Sam Altman and even Elon Musk, and we start to catch on to the idea that the three big AI companies are colluding to form a “cartel” of sorts. Amodei, Altman, and Musk may be repeating nebulous concerns about rogue agents and the apocalypse to possibly ice out competition from cheaper open-source models.

We realize that as Anthropic and OpenAI race towards their IPOs to mint new millionaires and billionaires, their need for cash is nakedly apparent. In Anthropic’s own words, they are profitable with the exception of the cost of training their models. In other words, what is literally needed for Claude to exist.

Now, some venture capitalists like Marc Andreessen and David Sacks are upset, because they believe we actually need more competition from startups—as well as to speed up innovation and “win against China.”

If all of this weren’t enough, Trump finally chimed in yesterday—predictably upset that he was not the center of the universe—calling AI’s destruction of humanity a “hoax,” and declaring that a “high IQ” person such as himself can handle any potential threat.

So what do we take away from all of this? Is AI very dangerous? Are OpenAI and Anthropic trying to form a duopoly, using their safety language to sidestep antitrust scrutiny? If we slow down, do we risk losing to China, whatever that means? What does “AI taking out humanity” actually look like in practice?

All of this—the many stakeholders, the conflicting motivations, the swirling narratives—are enough to make anyone’s head spin, myself included. At the same time, the narrative may actually be more straightforward than we think.

The most compelling argument for AI “destroying humanity” that I’ve seen actually comes from a documentary called Ghost In The Machine. The documentary, now out on PBS, argues that AI technology is actually a cover for precipitating a more colonialist agenda—one where the military fuses with technological advancements to enrich a small subset of people and extract from the rest. The people on top anoint themselves as “geniuses,” even men “giving birth.” And using some of the past language of eugenics, they dominate and exploit in ways that remind me more of the past than the future.

In this context, Trump’s usage of the term “IQ” is interesting, because IQ tests were pioneered by the founders of statistics and machine learning—a sort of pattern recognition that LLMs use today to respond to our prompts.

Subscribe now

The pioneers of machine learning, while lauded as rational and logical, are also people who had very apparent ties to eugenics and sterilization movements. Historically, the language around “IQ” and mathematical intelligence have been subverted to justify the enslavement and colonization of people who were ultimately just defenseless against people who had weapons.

With this as context, then the many distracting narratives about China and IPOs and Hugging Face may actually be a feature to the AI industry, not a bug. The focus on AI deflects from the reality that this is more about ideology than technology—the ideology being that a few people at the top truly believe that the rest of us should not have agency. The focus on AI also obscures the fact that human beings, at the end of the day, are still in control.

As Cory Doctorow writes, “The Hugging Face hack isn’t a mysterious, supernatural occurrence. It’s a Python loop and a chatbot. The people responsible didn’t accidentally create god: they created autonomous malicious software and then failed to closely monitor it, resulting in it doing something both foreseeable and bad.”

Now, I’m not saying that a swarm of AI agents taking down a 911 system is not a very bad and scary occurrence. It is. But there are ways to regulate and control products from being defective and causing harm without claiming they’ll end humanity. When people whistleblow about airplanes, for example, we don’t say that swarms of planes are going to lead to the destruction of our species.

Also, the hypothetical of AI models killing people…well, that actually isn’t a hypothetical. This has already happened via very brazen AI-military partnerships. As Dr. Heidy Khlaaf of the AI Institute writes on X, “A reminder that Anthropic has a partnership with Palantir, and their unreliable models are actively being used domestically and internationally to in fact, ‘kill a bunch of people.’”

Because all of this talk about the sci-fi future is a purposeful method to distract us from the present, whatever it is that we may be experiencing. If our present is harrowing, then sci-fi predictions deflect away from war, rising costs of healthcare and childcare, wildfires, boiling oceans, and people being poisoned by data centers. If our present is beautiful and joyful, sci-fi messaging injects a dose of fear to remind us we need these guys “in control.”

In an essay about our obsession with predictions—and how prediction-speak is more about control and power than profound clairvoyance or mathematical brilliance—Oxford philosopher Clarissa Velez writes that “The future exists only in our minds. To be more interested in it than in the present is to be more attached to one’s hopes and fears than to the world.”

Speaking of the world and the many beautiful things we can experience here, to place such an emphasis on “superintelligence” as the only form of intelligence also distracts us from our somatic senses, which our (very wise) nervous systems tap into through lived experiences.

This intelligence—emotional and relational—is something that cannot be understood by an AI, because AI is disembodied. As former big tech data scientist Abi Awomosu put it, “a model can generate fluent language about grief, soil, childbirth, danger, intimacy, rhythm, or belonging without possessing the embodied history through which a human nervous system learns these things. It models the descriptions of experience. It has not lived the experience.”

It’s true: we are intuitive creatures, more than just inputs and outputs. It’s both scary and wonderful. What real “AI danger” may come down to is “superintelligent” people tricking us into believing we’re not smart or rational enough to understand them. The true danger of AI, perhaps, is the deceptive messaging that leads us to believe we do not have agency or freedom in its system.

And yet, I found myself enthralled by Apple’s latest product launch event on Wednesday. I watched the Apple stream until it ended so I could see the company’s first attempt at a foldable phone—technology that already exists, by the way. My verdict: the iPhone Duo looks cool! It’s way too expensive, but all things equal, I’d probably purchase it.

I’ve realized that I am pining for consumer technology that is self-evidently useful and interesting. Even accounting for marketing tricks and the usual overhyping of Apple products, I think it’s nice when the executives of a tech company can easily explain what their product does and why you might like it. (The new Apple CEO is a former hardware engineer!) There’s comfort in embracing the naivety that a tech company might employ people who are at least somewhat driven by creativity and the user experience, instead of existentialism and unlimited power.

You can probably guess what I’m alluding to: the artificial intelligence industry. What am I supposed to make of the events of the last week or so? Nvidia purchased Hugging Face for $12.9 billion, which is a low-stakes bet on open models, I suppose. Meta launched Muse, a personal AI agent that doesn’t seem to work very well; despite assurances from the company’s entirely untrustworthy founder Mark Zuckerberg, Muse presents significant privacy issues. Nvidia CEO Jensen Huang wrote that “AGI has arrived,” referencing OpenAI’s latest and greatest model, GPT-6 Astra. It’s not the first time he’s declared AGI has arrived, nor will it be the last.

Then The Intercept started to pull back the curtain on the military contracts between the Pentagon and leading AI providers. A machine learning researcher who examined the (heavily redacted) contracts with Anthropic, OpenAI, Google, and xAI told The Intercept, “This indicates these labs and their personnel have been far more closely integrated into DOD AI development than previously understood and casts some doubt on the supposed contractual limits they have intended to place on the DoD’s use of these systems.”

Around the same time, The American Prospect published a disturbing report about how Anthropic is “building out an extensive monitoring system to keep tabs on activists who oppose the rapid development of artificial intelligence.”

The reports from The Intercept and The American Prospect were swallowed up by a genuine breakthrough and a peculiar X thread. The breakthrough: AI was apparently instrumental in (maybe? probably?) solving one of mankind’s most complex math problems, the sort of unimpeachable feat that makes me waver from my generally Ed Zitron-adjacent skepticism about AI’s viability. Except…OpenAI (maybe? probably?) swooped in to take credit from the researchers who were already using Codex, among other LLMs, to solve said math problem—thus proving again that Sam Altman is running around with his head cut off, desperate to demonstrate the potential of his models before he’s out of cash.

As for the X thread: a 27-year-old researcher named Jacob Coxon, who previously worked at OpenAI, announced he was quitting his most recent gig at Anthropic because “neither company is acting responsibly,” he wrote. “They are racing straight to self-improving superintelligence and gambling with our lives.”

These sort of warning shots are not new, and they typically do not go mega-viral. Coxon’s did. In ensuing interviews, he acknowledged that the timing of his announcement is a major reason why people are spooked—there are more and more documented cases of AI agents escaping their testing constraints, followed by frontier labs belatedly fessing up to these hacking incidents with a shoulder shrug. It did not help matters that another Anthropic researcher responded to Coxon’s thread with, “Jacob is correct here—we really do earnestly believe AI could kill all humans! I personally think it is >10% within the next decade. I believe Anthropic is trying its best, but we do not yet have a plan to solve alignment for superintelligence and are not clearly on track to.”

I’ve found Coxon’s ensuing media circuit to be underwhelming. He’s playing the same notes as other AI exiles. From what I can tell, he only worked at Anthropic for a few months. It’s difficult to ascertain how much he really knows about Anthropic’s inner-workings, and none of the interviews I’ve read and watched have clarified that crucial bit of context. This is not to say I trust Anthropic or its leadership team. I do not at all. Which is part of why I’d like harder evidence from Coxon about what he saw on the “inside,” rather than him offering incorrect assertions that Anthropic doesn’t leak anything, plus ridiculous comparisons about how humans are to monkeys what AI is to humans.

Coxon’s vague and dire warnings, coupled with chillingly blasé Kalshi-style predictions about whether AI will usher in the apocalypse, have mostly resulted in everyone scurrying back to their corners, just with more gusto. Sen. Bernie Sanders is (still) pushing a ban on the development of AI superintelligence, a bill that has zero chance of passage given current congressional ratios. Democratic presidential hopefuls pounced on Coxon’s thread but are suggesting solutions that can be summed up as “someone needs to DO something about this!” (Notable exceptions are Sen. Chris Van Hollen and Rep. Ro Khanna, who’ve both pitched more substantive ideas.)

As of Thursday afternoon, Altman and Dario Amodei haven’t said a word about Coxon’s X thread, or how employees at both of their companies are wondering aloud about the world ending because of AI. If precedent serves, both men will eventually wade in and say something uninspired about how more cooperation between frontier labs is necessary. They will not initiate any sort of cooperation, they will not acknowledge each other in public, nor will they consider a bilateral agreement with China, which is a necessity for any sort of superintelligence slowdown to stick. China will remain the bogeyman that Altman and Amodei point to while they race towards an IPO.

China is bad will be the modus operandi of other tech oligarchs and their MAGA allies, too. Sen. Ted Cruz put it plainly; he just told Politico the United States can’t slow down its AI development because China won’t either. “This is somewhat tongue-in-cheek, but it’s not entirely: if there are going to be killer robots, I’d rather they be American killer robots than Chinese killer robots,” he said.

Cruz went on to claim that Chinese AI firms have zero safety guardrails, and “it would be a terrible outcome if we stifle U.S. innovation and China pulls ahead.” What does it mean for China to “pull ahead?” You’ll rarely find someone from the deregulatory crowd define such outcomes. Nor will they explain why our AI industry must be fully privatized, or why any dialogue with China is forbidden.

Coxon’s thread has more than 155 million impressions as of Thursday afternoon. Conspiracies are popping up about his background and his motives; some, like Elon Musk, believe he was boosted by AI safety advocates. Maybe? Probably? I don’t think it was a nefarious effort in the way that Musk—a supervillain who changes his mind on AI regulation every week—is presuming.

As much as I’d like to trust the warnings of researchers like Coxon, I find it impossible to separate a slurry of competing interests, legitimate worries, and hubristic predictions about AI superintelligence. Some AI doomsayers would still risk their arbitrary estimates of global catastrophe if it means they remain eligible for a utopian outcome that just so happens to reward them with a bounty of riches. Others catastrophize on purpose, as a marketing ploy. Others won’t admit that their catastrophic predictions are largely predicated on frontier labs somehow penciling out and becoming relatively efficient—an outcome that remains elusive and hard for me to imagine, but is constantly glossed over by stakeholders.

What are we supposed to do with all of these conflicting signals? I don’t know. I really don’t. I despise how the AI industry and its thought leaders describe the world through abstractions. They aren’t just simplifying the stakes, or the underlying technology, to make it easier for laymen to follow along. When they blithely toss around odds about the destruction of humanity, or release cutesy GDP growth model scenarios, it’s because they still view AI development as a theoretical exercise. That sort of cognitive dissonance is very convenient; it’s a great fallback whenever another whistleblower emerges or a horrifying, real-world outcome can be attributed to an LLM.

My best guess is it would take a complete repudiation of the Trump Administration’s corrupt relationship with Big Tech, or OpenAI and/or Anthropic going bankrupt, to rein in and humble the AI industry. Here’s hoping. But if we can’t get AI frontier labs to cooperate—or for the United States and China to work together on safety precautions—then can we at least get one goddamned decent, affordable, tangible AI-powered consumer product before the world ends?

Here’s what else we’re reading this week:

• Alex Heath, a famous tech reporter who recently landed impressive (and well-done) interviews with Sam Altman and Mark Zuckerberg, is joining a venture capital firm. He also says he’s going to keep running his newsletter, Sources. I am not especially precious about the obvious conflicts of interest this will present; fashion magazines are a cesspool of spon-con, and Substack has a bunch of newsletters that are more marketing than they are journalism. All I ask is that Heath not insult our intelligence by claiming that he’ll run Sources “independently” of his work for the venture capital firm.

• I enjoyed this retelling of a somewhat infamous private meeting between Marc Andreessen and two Biden-era AI advisers, the latter of whom wrote for Politico that Andreessen is full of it, basically. From their write-up: “We most certainly did not say there was a plan—secret or otherwise—to classify the math underlying AI. Nor would it be possible to do so, since the core math of AI, linear algebra, is taught in high schools.”

• Here is a photo of Palantir CEO Alex Karp wearing a keffiyeh.

They don’t name the companies, for obvious reasons — the same reasons “Lucas,” their source who is also part of a unionization effort within his own tech company, is not named Lucas. Firing a worker for trying to unionize is a federal crime, but it’s not news that it happens all the time.

It happened to Clarissa Redwine.

She and her colleague JS Tan co-authored a book —Against Tech Oligarchy, releasing this Tuesday — detailing the history of labor unions in tech, telling their personal stories from Kickstarter, Microsoft, and Twitter, among others, analyzing what the movement has done right and wrong so far, and laying out the stakes and necessity of its persistence. They have both been visible faces and names in tech organizing this decade. Both experienced slow-building catalysts in their early tech careers that led to them confronting the gap between the industry they thought they were entering, and the one they were actually contributing to.

In 2020, Kickstarter became the first white-collar tech workforce in American history to successfully unionize. The summer before, at the height of the company’s union-busting efforts, management fired three of the most visible union advocates and organizers. Redwine was one of them, and was fired right after her most successful quarter to date. The illegal retaliatory firings had the opposite of their intended effect, and won over some employees who had been on the fence about unionization. The next February, Redwine gathered inside the National Labor Relations Board’s Brooklyn office alongside her former colleagues to hold their breath as they watched the vote be tallied. When one of the lead organizers turned and held up his notebook with the tally, she leapt out of her chair.

“After I was fired for organizing, I became somewhat known in this tight-knit community of tech organizers as someone who’s been through the fire and experienced retaliation,” Redwine told me. “And once JS blew the whistle and wrote this piece about Microsoft accelerating climate change, he was experiencing some heat from that. And JS had been working on this archive Collective Action in Tech for a while, and invited me to contribute.”

Tan remembers the Kickstarter union as a visible turning point for tech workers organizing within their companies.

Share

JS Tan came out of undergrad in the mid-2010s with a computer science degree and belief that technology had great potential to be beneficial to society; he cited the democratic nature of social media as one example. The optimistic nature of tech’s culture of that era, they write, was intentionally appealing to its workers. Mottos like “Move fast and break things” and Google’s now infamously abandoned “Don’t be evil” set an industry ethos of upending the status quo in the name of building a better world.

Tan’s disillusionment with that culture had been building for a while during his time at Microsoft. “[Microsoft Cloud] was a division that was, for the most part, less talked about until recently,” he told me. “You know, in the 2010s, the focus was on these consumer apps, these platforms, social media, whereas I was working in the Cloud computing sector that was this quietly growing— and, arguably today, the most important part — of Microsoft, and this was the division that was the reason the company was contracting with ICE and the Pentagon and Chevron.”

A moment of clarity he writes about in the book came in 2019 when the company sent him to Kazakhstan to present Microsoft’s newest AI and cloud technologies, and how they could be used to optimize oil extraction, to a team from Chevron. That same week, a collective of tech workers within Amazon launched an open letter demanding the company take urgent and assertive action on climate change. Tan found himself staring at the juxtaposition of the public stand from Amazon Employees for Climate Justice, and the presentation he was preparing.

That December, Tan published an essay about his experience in Kazakhstan, detailing his industry and employer’s role in accelerating climate change. It made his inner conflict public: the industry’s branding as a force for good was becoming increasingly impossible to reconcile with the reality of who ran it.

“Even before I was sent out to Kazakhstan, I was already seeing these various collaborations take place and was already participating in organizing within the company,” Tan continued. “I think like so many others, we felt this sense of betrayal, to some extent. This kind of, you know, what we’re actually doing at the company is not the kind of vision of tech that was portrayed to us in college.”

A couple years and a lot of work later, Redwine and Tan were hired by a coalition of workers within Twitter who were alarmed and horrified by their platform’s role in the January 6th insurrection. Employees pushed then-CEO Jack Dorsey to remove President Trump’s account after he’d used it to disseminate instructions to the rioters in real time. Dorsey only agreed to do so, with reluctance, after hundreds of Twitter’s employees signed an open letter that bluntly stated, in part, “We helped fuel the deadly events of January 6th.” The momentum from the success of the letter surged employees towards unionization.

Months later, as Redwine and Tan’s work with employees there was beginning, Twitter was bought by Elon Musk. The subsequent slashing and gutting of the company and platform as we’d known it quashed all momentum.

Tan and Redwine say neither Twitter employees’ activism nor Musk’s takeover were hermetic acts. Momentum had been building among tech’s working class for a decade, as recognition built that they were laborers with considerable power. Retaliatory suppression from its top executives had been sharpening into standard, preventative procedure.

Tech is the most powerful and consequential industry in the world. It’s in the better interest of its billionaires for their workers to not consider their weight in its power. Tan and Redwine want them to remember, even when a battle like Twitter’s is lost, that the power of workers is substantial, the fight remains winnable, and that redistributing some of the power tech oligarchs have hoarded is the service to society many disillusioned tech workers once believed was the altruistic missions of their companies.

“Even before my time at Kickstarter, I was interested in startups because it felt like, you know, you’re part of making the world a better place,” Redwine reflected. “You’re part of this cradle of innovation, and innovation is synonymous with progress, which, I don’t feel that way today. The internal mantra that one of the [Kickstarter] founders came up with was ‘Fuck the monoculture’. We had t-shirts of it. I really bought into that ethos. And it took me a long time, and it took a lot of clashes with the boss at Kickstarter, for me to truly internalize class consciousness and realize I was on the other side and could be expendable at any moment.”

“We started writing this book at a time where we thought that the tech worker movement was really in this lull,” said Tan. “But I think since then, and especially since we submitted our draft, there’s been kind of a bunch of interesting collective actions in the tech sector. And, you know, I think people are kind of becoming interested again in the possibility of tech worker activism and sort of the potential that tech workers have in sort of holding this industry accountable.”

Tan and Redwine analyze the evolving tactics of tech organizing in the past decade, and reflect on strategies they themselves used and perpetuated that are no longer viable today.

“I do think that activism within the tech industry is still important, because workers clashing with the bosses, is showing to the public and to the world that it’s not that the world should be against tech at large, but rather it’s that there is a small group of people within the tech sector,” said Tan. “The bosses, the oligarchs, have particular interests that differ from the rest of us. The second reason it’s still important is, even if tech workers are not winning something for themselves in this first action, the very process of participating in collective actions in a movement where they’re fighting against their bosses is a very revealing moment for them.The tech industry is this very ideological place. When we started in the industry, it was very ideological for very different reasons. And it’s confusing to untangle the class relations in such an industry. I agree with everything Clarissa is saying on the limits of writing an open letter in this, you know, in the mid-2020s. But even that experience of writing an open letter and seeing how the boss responds is going to be revealing to a lot of tech workers in having this sense of, ‘us as workers have these sets of interests, and our bosses over there have these other interests.’ So, you know, in terms of raising class consciousness, I still think it is important.”

There is no full set of uniform instructions for unionizing a tech company. Organizing in tech looks both the same and different as organizing in other sectors of labor. The class-rooted issues are fundamentally the same. In tech, however — especially in the age of AI — tactics like strikes have to be strategized and structured more intentionally. The function of many tech jobs is to design a product that can eventually work without the people who built it that way. If unionization in tech culminates in a strike, it’s not going to look like walking off a job site.

“Something that we stress in the conclusion of the book is the process of workers discovering how to seize the means of production, how to actually wield their leverage as workers at a firm — this is something that every group of workers that has had a successful strike has gone through. And tech workers are no different in that regard. We need to test, and we need to see how the boss responds so we can learn how our strikes cannot be undermined in advance, and what strategies work. And that may look different across a tech company in different departments. Striking might look different. We’re really inviting tech workers into the challenge of figuring out how to strike effectively, which workers in other industries throughout history have had to do the same work: of thinking critically about their labor process, how their company functions, and how a strike could be effective. And we’re hoping tech workers can discover that.”

If all of that weren’t strange enough, there was also a very specific and niche ethos that appeared to unite almost everyone in using burner phones and searching databases of riders. That was a deep hatred for the taxi industry.

Now, I can understand that some people don’t particularly like taxis—that they could plausibly find them less accountable, less accessible, or less comfortable. That’s a totally valid consumer preference.

I can also recognize, on a psychological level, that my coworkers probably had to adopt this sort of viciousness against something, in order to feel like they were fighting for something. Forcing Uber into being was a blunt task. And I think for a lot of people, the motivating mission couldn’t just be money, or allowing people to roll like a potato from their beds into a car. There had to be some kind of world-improving reason for the change, at least in their own minds.

But none of this ever warranted, in my opinion, making hatred of the taxi industry the kernel of one’s identity. It seemed like a strange hill to die on.

Well, it turns out that in 2026, it is no longer a hill to die on. Rather unsurprisingly, Uber and its people may not hate taxis so much—especially when it is explicitly in their interest not to.

In these times, Uber—recognizing that their autonomous vehicle production is behind, and that Waymo is eating up their ridesharing and delivery market share as a result—is now joining forces with the taxi drivers. In a wild about-face, Uber is partnering with unions to advance multiple bills in places like New Jersey, which would mandate a certain number of human drivers on the streets alongside autonomous drivers. From a recent piece in the Financial Times:

The ride-hailing group that became notorious in its early years for skirting regulation and aggressively lobbying against worker protections has recently joined unions in pushing to limit the rollout of self-driving taxis in locations such as New Jersey and Washington DC.

Across the US, Uber has argued for a slower deployment of autonomous vehicles to ease the transition for drivers — and pushed for rules requiring ride-hailing platforms to have “hybrid networks” that combine self-driving cars with human drivers.

As part of this “hybrid network” push, Uber is really on the offensive; putting out statements admitting to the “irony” of it all, and even pushing reports that warn of driver wage depression without a hybrid human-AV set-up.

I probably don’t need to tell this group of readers that driver wages are not actually what Uber cares about. Case in point: while Uber is putting out reports about helping drivers earn more, they also pitched out this story last week: Uber beats Waymo to launch robotaxis in London. The first line of the piece is: “Uber beat out Waymo in launching a commercial robotaxi service in London, the city’s first.”

The dissonance is not going unnoticed. Last week, I came across a post written by Alison Taylor, a Professor of Corporate Ethics at NYU Stern School of Business. Taylor outlines the problems with Uber’s strategy from a reputational standpoint: “One is the public, who increasingly hate corporate hypocrisy and flip flopping more than anyone else. A bigger one is a huge history of mistrust, which makes cooperation with unions difficult.”

I agree with Taylor that the company is making a messaging mistake in pretending to care, especially with customers and partners. Taylor also commented on the business and labor strategy: “How many other businesses are going to belatedly realize workers are important, and start backing out of their current ‘strategies’?”

On that point, I do agree that there will be a course correction to hire human workers when corporations realize they have automated too far. But it is equally true that Uber’s CEO Dara Khosrowshahi is under pressure to automate the fleet quickly.

Just last week, the company laid off 10% of its workforce and mandated a return-to-office for 99% of their employees, claiming that this makes their operations “simpler and faster.” All efficiency jargon aside, their singular goal, of course, with all of these initiatives, is to maximize shareholder profit (Uber’s shares went up 2% after employees were laid off.) While Uber lays off staff and makes mid-term plans to partner with the taxi industry, it seems that in reality, the company is buying time to catch up with Waymo.

I find it important to understand who the people are behind these strategies, and what is motivating them (I suppose, no longer a hatred of taxis). As has been documented extensively, the company has worked for years to “mature” following the removal of its mercurial founder Travis Kalanick. That initiative has been steered, in large part, by a fixer named Jill Hazelbaker.

Hazelbaker, who currently has the title of President and Chief Corporate Affairs Officer at Uber, worked for John McCain, Google, and Snapchat before joining Uber. For a time, she worked under former Uber, Google, and Netflix comms boss Rachel Whetstone; Whetstone then left and Hazelbaker took over. (It’s worth noting that Whetstone’s husband is seeking political office, drawing very recent comparisons to Donald Trump. I recall sitting at a dinner with Whetstone once where she cackled about drivers in India protesting fare cuts, by carrying the corpse of a driver who took his own life to the Uber offices.)

I never saw Hazelbaker cackle in such an explicit manner (though, under the guise of offering me a job at Snapchat, she did fly me to Venice Beach in 2015, proceeded to ask me probing questions about Uber’s org chart, and subsequently joined Uber to lead the team the next month while pretending not to know me.)

In any case, while Uber has had less public relations crises under Hazelbaker than in the past, it can only hide behind a facade of “maturation” or civility for so long. Especially when the actual core of its ethos may not actually be disdain for taxis, but more predictable corporate motivators: self-interest, survival-of-the-fittest, and the need to maximize shareholder value and executive compensation.

Some of that true ethos bubbled to the fore in this New York Times report documenting how Uber fights sexual assault lawsuits. In one deposition, an Uber lawyer demanded to know what a rape victim was wearing the night of her assault, asking whether she ever felt “abandoned by her father or degraded by her mother.” The vile tactics of victim-blaming aside, such a lawyer and line of questioning reminds me of what a corporation will actually stoop to to “win” at literally all costs.

While conducting some research for this piece, I happened to see that Uber’s former head of public policy is now the head of global public policy at Waymo—presumably now fighting a government relations war against some of his former Uber colleagues, who have now banded together with their prior enemy, the taxi industry.

Amid this tangled web of allies and enemies, it may all feel disturbingly fluid, and it all may feel like a game meant to benefit a few people. But it’s also a game in which real people can become collateral damage. In this next phase of Uber, a dark side they may have once masked in the pursuit of “maturity” is now becoming more brazen. We’ll see how customers and regulators respond.

What else we’re paying attention to this week…

The reviews are in for Alex Gibney’s scathing documentary about Elon Musk: “The film depicts the world’s richest man as petty, narcissistic, megalomaniacal, cruel, detached from humans at a personal level (including various ex-partners and many of his children) while purporting to want to save humanity by colonizing Mars. A real prince. Musk, in my view, gives Elon the takedown he richly deserves.”

People in Kenya made money ghostwriting college essays and other homework for lazy high school students. When AI arrived, the cottage industry went bust.

A campaign to support the Ellison-Paramount merger is literally putting words of support for the merger in writers’ mouths.

Data center “enthusiasts” and proponents are leveraging an abstract fear about China winning the AI race to push for more centers.

A delightful piece of news, for a change: after getting laid off from a tech startup three years ago, a woman and her partner opened the first Indonesian coffee shop in Silicon Valley.

Subscribe now

The title and subtitle give a fair sense of Karma’s thesis:

Karma’s argument can be summarized as follows:

1. AI poses catastrophic risks.

2. Slowing or stopping data centers from being built in particular communities may address “local environmental concerns” but would not address the catastrophic risks.

3. Therefore, anti-data-center efforts are a “distraction.” People should instead build a global movement for AI nonproliferation, akin to the nuclear nonproliferation movement.

But (1) is a dubious statement and (2) ignores the very-much-not-dubious concerns driving grassroots opposition to data centers so (3) ends up making Karma sound like a condescending, out-of-touch elitist who has spent lots of time absorbing AI hype but little (if any) trying to understand and empathize with the concerns underpinning the anti-data-center movement.

To keep this piece reasonably short, I’ll focus on a few particular lines from Karma’s piece that stood out to me when I read them.

The first half of the piece is an attempt to convince people that AI poses apocalyptic-level risks. That starts with the article’s opening:

America’s biggest AI companies have spent the past few years churning out advanced new models as fast as they possibly can. Now the industry is screaming out with a nearly unified voice: Please, for the love of God, make us slow down.

That is a fair, if hyperbolic, characterization of some recent statements by AI companies’ executives. But it is foolish to take those statements at face value. The track record of Big Tech and venture-capital-backed companies is not that of people who care about the public interest or laws designed to protect it. Their record is instead one of particularly aggressive profit-maximizing corporations who show, at best, callous and reckless disregard for the harmful effects of their products, if not a conscious desire to inflict those harms, and view the law as something to be, in order of preference, exploited, ignored, or attacked.1

A more plausible interpretation of the recent “slow us down” statements would be: “We really need to attract money from investors and lenders, and, right now, rising public anxiety about AI means that an effective way of doing that is to make our models sound super powerful and scary. Please, for the love of God, signal-boost our scare mongering so that we can get more money.” Karma duly obliged.

A few lines later, Karma decided to engage in some hype-fueled AI scare mongering of his own:

Plenty of others in the industry have warned that we’ve already crossed a threshold from which there is no turning back. They worry that the most dystopian fears of an AI future—chatbot-generated bioweapons, major hacking attacks on banks or governments, rogue AI agents wreaking havoc on society—could soon become regular occurrences.

Leaving aside for the moment what exactly a “chatbot-generated bioweapon” is…

…actually no, let’s not leave that aside. Karma did not provide a link to explain what he meant by “chatbot-generated bioweapon,” so it’s unclear what exactly he was going for other than “I want to make AI risks sound as scary as possible, and I need to come up with something scarier-sounding than automated cyberattacks and more concrete than ‘rogue AI agents wreaking havoc on society.’” “Chatbot-generated-bioweapon” certainly sounds concrete and scary, but the hard part about building bioweapons has never been that no one knows how to formulate or even produce one—dark web marketplaces to obtain the components for bioweapons have been around for years.2 The hard part is getting the money, materials, equipment, and array of trained experts necessary to actually produce them in the physical world.

That points to the deeper flaw in Karma’s scare-mongering: all of the harms he mentioned are not really risks associated with AI, but with human actors. LLMs do not think, understand, or have intentions. They do not, on their own, decide to send phishing emails, launch cyberattacks (much less bioweapons), or go “rogue.” It’s true that the cybersecurity landscape in the coming years will be hellish because LLMs are great at helping humans launch powerful cyberattacks but not so great at helping them write secure code. But like other software used in cyberattacks, LLMs are only capable of doing dangerous things if there are malicious, or at least grossly negligent, humans that direct it to do those dangerous things and/or fail to put reasonable safeguards in place to stop them.

Subscribe now

That is what what happened with recent incidents (including the UK AI Security Institute incident that Karma mentioned) where bots3 supposedly “broke containment” and hacked third parties. Despite the credulous media coverage that anthropomorphized the bots, the failures in each instance were thoroughly human. In each case, humans directed the bots to find and exploit security vulnerabilities while providing them an environment that contained security vulnerabilities. In the OpenAI “hack” of Hugging Face that grabbed so many media headlines, the researchers also did a blatantly awful job of monitoring the bots. Indeed, the weakness of the safeguards makes it more than plausible that Anthropic and OpenAI were low-key hoping the bots would break “containment” so they could get the resulting media attention from a hack. Which, of course, they did.4

That leads us back to why Karma wrote this piece in the first place. His thesis is that countries need to pass “an AI non-proliferation treaty” to address these catastrophic risks and that merely focusing on restricting data center construction is thus a (to use the word from his headline) “distraction.”

He elaborates on this point further down in the piece:

Unfortunately, the main form of AI-related legislation being proposed at the moment would not get the job done. State-level data-center moratoriums, such as those enacted recently in New York and Texas, might slow down AI companies a little bit by disrupting their construction plans, but they will eventually find other places to build. Even if Congress decided to pass a federal ban, AI companies could just set up data centers in other countries.

The point slapped Karma in the face and he still missed it.

Data center protestors are not stupid. Of course they realize that if they stop data centers from being built where they live, the AI companies will just build the data centers elsewhere. But making them find other places to build is the whole reason for the grassroots opposition to data center construction. The people showing up to their local council meetings to protest data centers are not there because they want to disrupt the coming of the robot apocalypse. They are trying to prevent water-guzzling, cost-hiking, grid-straining, noisy, polluting facilities from being built in their proverbial and often literal backyards.

A Gallup poll from this spring asked Americans whether they would support or oppose the construction of a data center in their area. 61% of respondents said they would oppose it. Those 61% were then asked to provide some of the reasons for their opposition. The most-mentioned reasons related to effects on resources (50%), quality-of-life concerns (22%), effect on costs (20%), and pollution (16%). Only 4% mentioned “Scared of AI/Don’t trust it/AI will take over,” which would seem to be the appropriate category for “robot apocalypse.”

At the very end of the piece, Karma seemed to acknowledge that more immediate concerns were the animating force behind data center development. But rather than empathizing with those concerns, he demeaned them:

Meanwhile, according to the latest polling, 75 percent of Americans, including the majority of Republicans, oppose data centers being built near them. More than 500 localities across the country have either restricted or banned them. Could this energy be channeled into a more sweeping—and more effective—framework for AI regulation? So far, the data-center backlash has mostly focused on local environmental concerns, not great-power diplomacy. But then again, that’s exactly how the anti-nuclear movement began.

First, the anti-nuclear movement did not start over “local environmental concerns.” Concern about nuclear technology started after many people saw the most powerful weapons in history kill over 150,000 people at the end of World War II. The first significant anti-nuclear protests occurred after 23 Japanese fishermen suffered radiation poisoning after a U.S. nuclear test. The first anti-nuclear organizations and truly mass protests were pacifistic in nature, focused on nuclear war and weapons testing, not “local environmental concerns.” The latter only became part of the milieu later, after civilian nuclear power plants started popping up.

More importantly, however, the condescension in this final passage is unbearable. “Local environmental concerns” makes it sound protesters are worried about whether data centers will block their view of a local creek rather than cost of living, noise and air pollution, and access to essential resources. The overall tenor of the last paragraph, and really the piece as a whole, is: “It’s too bad all these peasants are so focused on trivial concerns like the quality of life in their community. But maybe their energy can be useful if it can be ‘channeled’ into a crusade to stop chatbot-generated bioweapons.”

The fact that Karma views the data center fight as a distraction from stopping chatbot-generated bioweapons, rather than the other way around, is a sad illustration of just how out-of-touch the AI-pilled intelligentsia has become.

Matt Scherer is a fellow at the Open Markets Institute, where his research and advocacy focus on developing policy responses to the eventual bursting of the AI bubble. His Hard Reset pieces focus on highlighting the risks posed by the AI bubble and pushing back against the hype that is inflating it. The opinions expressed here are solely his own.

So said Mark Cuban, the former owner of the Dallas Mavericks and longtime co-host of the TV show “Shark Tank,” exactly one year ago. Cuban was defending Steve Ballmer, an early-days Microsoft executive who later served as the company’s CEO and has spent his golden years as the heavily involved owner of the Los Angeles Clippers.

At issue: last September, investigative journalist Pablo Torre reported that Clippers star Kawhi Leonard had signed a “no-show” endorsement deal with a now-defunct (and fraudulent!) company as a means of circumventing the NBA’s salary cap rules—and that the deal was funded by Ballmer.

Torre’s report kicked off a nonstop sports media news cycle, and was damning enough that the NBA hired a law firm to investigate the matter. The law firm and the NBA released their findings this week. The top-line result is that the Clippers “violated the circumvention rules in numerous independent ways.” And: Ballmer “knowingly sought to help Mr. Leonard obtain off-court income opportunities and, in at least one instance, engaged in a significant act of team facilitation. Mr. Ballmer also failed to create conditions under which his organization abided by the NBA’s circumvention rules.”

Among the many penalties levied by the NBA: Ballmer is suspended from all league activities for a year, and the franchise must forfeit five future draft picks (and $30 million).

The NBA’s salary cap purposely limits how much teams are able to pay players. It’s mostly a leverage point for cheap-o franchise owners, who generally do not want to pay players what they are actually “worth” relative to their marketing value. To his credit, Ballmer is not “cheap” when it comes to player contracts or building an arena. Since purchasing the Clippers in 2014, Ballmer has feverishly tried to turn around the fortunes of the oft-hapless sports franchise. He privately financed the Clippers’ $2 billion Intuit Dome, for instance, which opened in 2024. (He’s cheap in other rich-guy ways, including how he reportedly takes advantage of tax write-offs, but I digress.)

It helps that Ballmer is unbelievably wealthy even among his team owner peers. Forbes lists his net worth at $156.7 billion, which makes him the ninth-richest person on earth as of Thursday afternoon. A little (alleged) cap circumvention to secure the services of Leonard, one of the 75 greatest players in NBA history, would be a small price to pay. Unless, of course, Ballmer and/or the Clippers get caught.

Ballmer and/or the Clippers got caught. According to Torre’s reporting (much of which was substantiated by the NBA’s investigation), after Leonard signed with Los Angeles in 2019, his uncle slash agent Dennis Robertson tried to pocket tens of millions of dollars via shady “endorsement agreements” facilitated by the Clippers.

Ballmer pleaded his innocence over the course of the last year, and he found a credulous audience. Up until the release of the NBA’s independent investigation, countless pundits and reporters strayed outside of their usual coverage areas and butchered their analysis of Torre’s scoops. Unable to contextualize financial reporting, and seduced by what one can only reasonably deduce were Clippers sources eager to muddy the waters, these pundits and reporters repeatedly cited a lack of a “smoking gun.” This is rarely how reporting works. Investigative journalism is not the equivalent of a court of law, and it’s unusual for a story (or even a series of stories) to answer every single outstanding question a reader might raise.

Torre, though, came pretty darn close. As time went on, and Torre peeled back more layers of the onion, the same refrains grew louder and louder. The cynical take, which I was more sympathetic to, was that the NBA wouldn’t bring the hammer down on its richest owner even if he did everything he was accused of doing. But the other common take drove me nuts: that by virtue of his immense wealth and entrepreneurial spirit, Ballmer was too brilliant to rashly participate in (or even have knowledge of) a cap circumvention scheme. Cuban asserted as much multiple times. A sampling of other equally annoying (and wrong) assessments:

• The Ringer’s Bill Simmons said, “This was so sloppy that it almost makes me think, I kind of just can’t believe—everything I know about Ballmer, how shrewd of a businessman he is, how thoughtful he is about everything… I just can’t believe he was this dumb. That’s kind of where I’ve landed. And maybe he was this dumb. But I can’t believe he was this dumb.” (Disclaimer: I was a Grantland intern and used to play pickup basketball with Simmons, who I did not know well but liked interpersonally.)

• An anonymous front office executive told The Athletic’s Jason Lloyd, “Steve is really intelligent. I don’t believe he’s dumb enough to wire $50 million to a company, just to have that company turn around and give it to Kawhi. I don’t believe it.”

• “It’s not something I would ever believe,” former Milwaukee Bucks owner Marc Lasry told CNBC. He added that Ballmer is “honorable” and a gentleman.

• “The Clippers are not dumb to first of all, leave a messy paper trail that [would] basically ruin their franchise,” ESPN analyst Bobby Marks said last year. “We’ll wait and see what comes out of this investigation. I’m not anticipating anything. That’s not any disrespect to Pablo’s reporting. But I’ve been part of salary cap circumvention cases. The optics don’t look good, but it doesn’t meet the smell test as far as a team circumventing the salary cap.”

Other reporters, like Puck’s Chief Snark Critics Peter Hamby and Dylan Byers, de facto backed Ballmer by insinuating that Torre had somehow screwed up his investigation. Disgraced sportswriter Jay Mariotti suggested that Torre should “quit… until he decides if he ever can validate a story.” These reporters did not offer much in the way of evidence—just that the NBA seemed to be taking a while to wrap up its own investigation, and that some poorly-sourced stories published by ESPN made Ballmer look like he was in the clear.

They were badly mistaken. The full 35-page report from the NBA and the law firm it contracted is quite detailed; it includes key witnesses, text messages, and emails that go far beyond “bad optics,” as Marks put it last year. You could argue that it, too, does not have a “smoking gun,” but again: what does that really mean? Lengthy, written investigations require interested readers to use their brains and make connections. No reasonable person is going to read the NBA’s report and conclude that Torre, or the league, were picking on the wrong guy.

Ballmer is apparently “fuming” about the penalties enacted by the league. “I know Steve Ballmer a little bit, and I understand when you’re worth $150 billion and you’re angry—and he’s very clearly angry—you can’t advise him, you can’t talk to him,” said ESPN analyst Brian Windhorst. The Clippers released a disappointed statement with a typo in it, and also sent a litigious letter to NBA Commissioner Adam Silver. From what I’ve gathered, Ballmer and his basketball team don’t have many avenues to actually fight the penalties assessed against them, though I’m sure that won’t stop them from trying. As Windhorst noted, hell hath no fury like a tech oligarch who doesn’t get his way.

The thing is, Ballmer’s behavior got the Clippers into this mess. Following up a self-inflicted scandal with a prolonged temper tantrum will only make him look worse, so that’s exactly what I expect him to do. Unlike Cuban, I don’t know whether Ballmer is “dumb” or not. I do know this: Ballmer’s cartoonish ruthlessness is hardly uncommon among tech oligarchs—and it absolutely should not be mistaken for intelligence.

Here’s what else we’re reading this week:

• For the second time in a year, a federal judge ruled that Google is running an illegal monopoly—and then declined to do much of anything about it. Judge Leonie M. Brinkema, who’s 82 years old and was appointed by President Bill Clinton, “ruled on Wednesday that Google must make changes to address its advertising technology monopoly but would not need to break up that business,” the New York Times reported.

  • Specific details about those “changes” are under seal for a few more weeks, but as The American Prospect’s David Dayen noted, Google doesn’t seem very concerned. Quite the opposite, in fact: “We’re very pleased the court rejected the DOJ’s proposal to break apart tools that help small businesses reach new customers and grow,” said Google’s vice president of regulatory affairs.

  • Last September, U.S. District Judge Amit Mehta—appointed by President Barack Obama—decided against breaking up Google, despite ruling that the tech conglomerate also had a monopoly over the online search market.

• Uber laid off 10% of its staff on Wednesday, which reportedly amounts to roughly 3,300 people. Like Xbox’s mass layoffs announcement in July, Uber is relying on vague language about how it needs to “remove layers” of management for the long-term health of the business.

  • Uber is also immediately shutting down operations in Nigeria and Uganda. The ride-sharing company is apparently offering what it calls a “goodwill payment” to active drivers, though it’s unclear how much money that entails, and it’s hardly a consolation for people who made a living off the app. “I cannot accept it. What do they expect me and my family to do? How do they expect us to feed?” one Uber driver said to the outlet TechCabal.

• Another week, another inexplicable Sam Altman quote. This one is from his recent interview with Time and Alex Heath: “I think everyone should have a personal robot,” Altman said. “I would love to have a personal robot that could do the tasks I don’t want to do. That’d be great.” In the same interview, Altman also unsubtly critiqued Meta’s pervert glasses. “I find it very uncomfortable to talk to people with a camera and a light,” he (correctly) stated, when asked about AI glasses more generally.

• There’s a four-hour Elon Musk documentary in the works, according to a feature from The Hollywood Reporter. It’s directed by documentarian Alex Gibney, who was unable to secure access with Musk himself, but seems to have locked down interviews with dozens of other important characters in Musk’s orbit. “People assume because he’s very rich that he’s not crazy,” Gibney said of Musk. “But the terrifying thing might be that he’s crazy and he’s very rich and powerful. That’s the warning bell the film sounds.” Assuming there’s no chicanery, the documentary will reportedly be released on HBO during the first half of 2027.

• Commerce Secretary Howard Lutnick is playing nice with Anthropic. Coincidentally, Anthropic co-founder Tom Brown had wonderful things to say about the Trump Administration on Wednesday. “I really love Trump’s post from earlier this week ... where he was pointing out that the data centers are just an enormous source of prosperity,” Brown said at a G20 panel.

Sam Altman: A few years ago, every community wanted us to come build data centers in their community. It was going to reindustrialize, it was going to bring these jobs, it was like, ‘yes, please do it.’ And then that really changed, and now people don’t. And I have a lot of empathy to why people don’t. The good news is, we have a very big country. There’s a lot of places with not many people around, and land that people don’t have much use for, and we can go put data centers there. So I think we can solve this problem, and I think we should be respectful of what communities want.

Alex Heath: It sounds like you think the general answer is putting the data centers where people are not.

Sam Altman: I think that’s a perfectly good answer… We can kind of put the data centers in a lot of places, and move the results around the world very easily, so we have more flexibility with this.

While I was back in Wisconsin this August, I traveled to Wisconsin Rapids and Port Washington to meet with residents who have formed opposition groups to the data centers being built or planned in their towns. Those two places — each with fewer than twenty thousand residents — are very different parts of Wisconsin. Rapids is an industrial town in the middle of the state, where a substantial percentage of residents are struggling financially. Port is a well-off enclave on the lush coast of Lake Michigan, surrounded by hiking trails and farmland. In both, an overwhelming majority of residents are desperate to stop their data centers. The stakes of one moving in are high and all-consuming. In Port, where Vantage, Oracle, and OpenAI have already broken ground, a neighbor of the new center told me their future had been stolen.

In both cases, it wasn’t public opposition right out of the gate. It began with clarifying questions about how a data center would affect their lives and livelihoods; how much more they’d pay in utility bills, how loud a functioning center would be at all hours, information that would be available had environmental impact reports not been strategically avoided.

They were ignored at best. At worst, they’ve been laughed at, arrested, and threatened.

I’d known the facts of what building a data center means to its local environment, economy, and community before I started this project. Meeting the people whose lives, homes, and futures are inextricably linked to these places has removed the abstraction from the fact sheets about data centers. Living next to one seems cruel and impossible.

Share

Altman and OpenAI obviously aren’t a monolith in building data centers, but through a highly visible relationship with President Trump and cover story interviews, he’s become a face of their development. His apparent surprise at the public’s change in attitude about data centers and AI aside — which occurred as the public learned what a data center is and what it does — I’m not surprised his PR strategy is to feign sympathy and promise he only meant to revitalize local communities as a conveniently positive side effect to AI progress. And if that’s not what you want, then fine. He’ll find plenty of land “people don’t have much use for,” in the US or not, to build these centers on.

There’s something in the way he chose to phrase that that sums up this new wave of colonial extractivism from Big Tech. It’s not a place people do live, work, farm, hike, raise families. It’s not an environment that would be forever changed and destroyed by the presence of a data center, with inevitable ripple effects across a region and watershed likely to exacerbate the climate crisis that will affect people, whether they’re next to the center or not. It’s unused land because nobody is currently deriving profit from it.

(It’s also just probably not true. Data centers don’t require thousands of workers, but they do need skilled workers. Tech is famous for building infrastructure in alluring areas to attract top prospects.)

Leave a comment

In Wisconsin, broadly and generally and despite the differences in the locales targeted, data centers will hurt the people who live there or the environment the state has protected fiercely throughout its history. The companies fighting to build them promise reinvestment and reinvigoration of the local economies and communities. That’s constantly contradicted by their actions and messaging within these communities. They aren’t hiring locally for the few jobs a functioning data center would bring — in some cases, they aren’t even hiring construction workers locally. “Closed-loop” systems that are posited as water-efficient, renewable methods for running data centers, are more of a PR buzzword than a peer-reviewed renewable alternative. Meta’s data center in Cheyenne, Wyoming, that used a closed-loop cooling system, circulated contaminated wastewater back into the city’s water supply after the fill and flush process. The data center contaminated Cheyenne’s water with a mystery bacteria.

Even before President Trump signed the Stargate Project into law on the second day back in office — with Sam Altman standing over his shoulder — there has been a pattern of local, state, and federal governments tearing down red tape to clear the way for data centers to be built before regulation could keep up. Any concerns about labor, the environment, or the safety of local residents can’t be answered, because the work required to find out has been intentionally avoided. We know they are not comparable, in terms of energy or water use, to previous industrialization; we know that they are using resources on an unprecedented scale. We just don’t know what that will mean down the road.

Share

As I’ve mentioned throughout these pieces, I’ve spent a good deal of my life in Wisconsin. Both sides of my family have roots there, and it’s a place for which I have a deep love, respect, and gratitude. That doesn’t make me hope that data centers are built out of sight and out of mind, away from the place I personally care about. Seeing this happen to a place I love has made it even harder to accept that somewhere else has to bear the consequences of AI infrastructure.

In Wisconsin, opposition to data centers began with questions. Before we decide where to put them, or if we should build them at all, those questions need to be answered.

As a result, this temp work can be prone to bad actors: for example, workers may actually end up getting less than a state’s or city’s minimum wage because staffing agencies with online platforms require the workers to pay for said platform to list availability for shifts. Many stadiums are not required to use union staff; therefore with fewer rules, staffing agencies that may skim wages and commit wage theft can slip through the cracks.

In Seattle, a partnership between Workers’ Lab, a non-profit building worker-centered tools, and the MLK Labor Council, a project of the AFL-CIO, sought to bring good jobs with livable wages to stadiums during the World Cup. Previously, organizers had formed a $1 million labor harmony agreement with the company that leases the stadiums for Seattle’s football team, the Seattle Seahawks, and the city’s soccer team, the Seattle Sounders.

According to Rigo Valdez, a decades-long labor organizer and executive director of the AFL-CIO’s MLK Labor and Worker Innovation Center, union workers at the stadium, since the agreement, have been making over $23 an hour—above the minimum wage in Seattle. Before the agreement, 17% of stadium workers were union workers; afterwards, 82% were.

Subscribe now

Ahead of the World Cup, leveraging FIFA coming into town, union representatives and workers’ rights advocates met with FIFA’s labor organizing representative to ensure that the values inherent to the union and the stadiums’ labor harmony agreement could be upheld and even expanded upon. The idea was, together with the community, to ensure that FIFA left something positive behind: that what happened at Qatar’s World Cup, where workers worked to literal death to prepare for the games, was not in any way replicated in Seattle.

“There needed to be a recognition that human rights include workers’ rights,” Valdez told me.

Valdez partnered with Adrian Haro and Catherine Huang from the Workers’ Lab, and together they created UNtemp, a digital platform that assigned shifts to workers so they could staff a bustling World Cup watch party at a mall near the stadium. Workers were responsible for crowd control and fan engagement, as well as food services, security staffing, and janitorial work.

As part of this initiative, all the workers who were not union workers joined the union—and together they formed a co-op, so that workers were owners in the output and could make higher wages, and so profit-sharing could occur if there were a profit.

UNtemp also was designed for workers to bid for shifts specifically in line with union rules, assigning people shifts based on seniority in the union as well as geography. (While other platforms like UNtemp do exist, the team who built UNtemp argues that those platforms not as customizable to the union rules.)

Huang and Haro reiterated to me that it wasn’t a foregone conclusion that the platform would be successful. This was an experiment. Could workers properly find and sign up for shifts? Did the platform fulfill 100% of the demand and appropriately match supply with demand? Did workers actually show up, and were they paid relatively seamlessly?

According to the stakeholders involved, the answers were yes.

Nearly 300 shifts were scheduled during the World Cup’s festivities, with 70 workers signed up for the platform and $32,000 in payments flowing through the platform to workers. While most workers were already part of the union, some users of the platform included non-union workers who then became union workers and W-2 employees—all with access to health, dental, and vision insurance.

There were no profits made from the co-op, but the co-op did break even—and, according to Valdez, workers also also felt a tremendous amount of pride and ownership in the co-op as something they generated and owned themselves. Workers even designed a logo for the co-op, called the Seattle Janitors Cooperative.

This is a stark contrast to working conditions in other cities, where unions were busted and World Cup work was cobbled together by temp work.

The experiment, while contained and relatively small in scale, made me think about how it’s also not inevitable for all gig platform work to be a race to the bottom on wages. Temp workers can provide good services for important cultural or sporting events while not feeling dispensable or disposable. And infrastructure decisions can be made to involve workers to share in the success of outcomes.

So, what’s next for UNtemp? The team at Workers’ Lab wants to scale their model to other stadiums and cities—potentially even to staff the 2028 Olympics in Los Angeles, and also to break through to construction sites in the South where workers and unions may face more of an uphill battle.

Valdez, of MLK Labor Council, is excited to see the app automate functions like payroll, which was handled this time manually through the employer—as well as expand to other stadiums in Seattle, and provide foreign language options for a workforce who may not be as familiar with English language interfaces.

“There’s a difference between us and for-profit platforms, because of our values,” Huang of Workers’ Lab told me. “We have a strong interest in learning, and sharing those learnings with the contemporaries in our field.”

Subscribe now

What else we’re paying attention to…

A Democratic candidate in Alaska is getting up there in the polls—and 20% of his contributions come from Anthropic tech workers.

Congress is increasingly concerned about AI’s ability to scan commercial databases to know whether Americans have gone to church or yoga.

Amid the AI boom, there is a “mansion shortage” in San Francisco.

A Bay Area tech worker is one of the many people missing following the devastating floods in Nepal.

Employees of startup Snyk, once valued at $8+ billion, are seeing a precipitous decline in their equity. “At Snyk’s peak valuation, their shares were worth over $10. By the summer of 2025, that had declined to around $3, the other former employee said. That figure is now down to $1.16 per share as of late August, according to an internal document.”

The gambit doesn’t work very often—or at least it’s not supposed to work—because the judges are (hopefully) impartial and not swayed by naked attempts at reversing the narrative of the match. These last-minute antics are built into the sport and considered to be fair game, but I’ve always found them unbecoming. You lost. It happens.

On Wednesday, Meta settled a long-running lawsuit with almost every state attorney general. The lawsuit alleged that Meta illegally collected data on children, that platforms like Facebook and Instagram are unsafe and addictive for adolescent users, and that there aren’t sufficient safeguards to protect adolescent users. By settling, Meta evades the scrutiny that would’ve resulted from a full trial, which began a week ago and was reportedly going very poorly for the defendants. Mark Zuckerberg, who looks like an animatronic figure at the Skateboarding Hall of Fame Museum, was spared of taking the stand. Meta is on the hook for significantly less in penalties—up to $16.7 billion—than it might’ve owed if it had gone the distance at trial. And technically, a settlement agreement means Meta doesn’t have to admit to wrongdoing.

But having read through the terms of the settlement, let me put it plainly: Meta lost. Which is why I find the corporation’s spin, its post-match prancing, annoying enough to be called out.

When news of the settlement broke, Meta released a statement titled, “Our Agreement With Bipartisan Attorneys General: Calling on TikTok and YouTube to Join Us in Supporting Teens.” Lots of curious word choices in the subject line. “Agreement” is similar but softer than “settlement,” for instance. “Bipartisan” suggests a harmonious alliance that includes Meta itself, rather than Meta’s conduct being so egregious that it inspired an adversarial coalition of plaintiffs rivaling Ronald Reagan’s 1980 electoral college wipeout. “Supporting Teens” is definitely one way of describing the proposals in the settlement. Another would be: “Significantly Limiting Teens’ Access to Our Platforms, But Only After Many Years of Us Skirting Accountability.”

The rest of Meta’s statement is more of the same revisionism. Like this:

“Over the years, we have consistently partnered with parents and experts—listening, learning, and building. That’s why we launched Teen Accounts in 2024, to bring automatic protections to teens, and more control for parents. Ensuring teens have a safe and productive experience on our platforms is an absolute imperative for Meta. We want to get this right for parents and teens, and that’s why we partnered with state attorneys general to set a new industry standard.”

Here’s what actually happened: “Teen Accounts” only launched after this lawsuit was filed, after whistleblowers came forward about Meta’s disregard for younger users, and after Zuckerberg was repeatedly flayed in front of Congress. No-brainer safety features for teenage users, like private-by-default Instagram accounts, weren’t instituted until after this lawsuit came about. Many other features that would hypothetically curb a user’s engagement time have historically been set as “opt-in,” rather than “opt-out,” and so they predictably have had little effect.

Hard Reset

New Court Filings Allege Depraved Internal Communications at Meta and Snapchat

The company line from social media platforms like Facebook, Instagram, Snapchat, TikTok, and YouTube is that they care deeply about the mental health and safety of their users. These platforms claim to be extra mindful of their younger users; they all boast about parental controls and other safety tools, which they say are intended to limit underage exp…

Read more

10 months ago · 264 likes · 27 comments · Alex Shultz

When Meta says it “partnered with state attorneys general to set a new industry standard,” it is, again, pushing a reductive framing. Meta was itching to duke it out against the state attorneys general; they are not happy-go-lucky friends. As a result of the settlement, Meta is on the hook for each of the following:

• Daily time limits of two hours for users under 18, unless the limit is lifted by an adult. The limit is cumulative across Meta platforms.

• All Meta platforms are blocked for users under 18 between the hours of midnight and 6 a.m.

• Notifications from Meta platforms are automatically muted from 8 a.m. to 3 p.m. during the school year.

• No more displaying likes or reactions for users under 18 on Meta platforms.

• No more “beauty” filters on Meta platforms.

• Prompts pop up after every 15 minutes of continuous usage on Meta platforms.

• Teenage users can switch to a non-algorithmic feed.

• Parents can gain access to more tools to monitor their child’s social media usage, including whether their child is trying to create “secret” second accounts.

There are plenty of details to iron out, and as part of the settlement, an independent auditing team will keep an eye on Meta’s compliance. It won’t be easy to authenticate who’s an adolescent user and who is not. And any age-verification measures present their own drawbacks, both in terms of user privacy and freedom of expression. I share those concerns and worry about their implications—I’m just not convinced there’s a great alternative for young users, who are, in my view, especially vulnerable to social media’s harms.

Do Zuckerberg and the other higher-ups at Meta feel similarly about these harms? With their own children, certainly. But Zuckerberg and his cronies have a business to run. Their primary goal, as demonstrated by the statement that Meta released on Wednesday, is to limit reputational damage and prevent stock price slippage. On the latter point, mission accomplished, at least for now. Meta investors seem to be relieved by the paltry financial penalty in the settlement (the corporation’s market cap is, after all, $1.45 trillion), and that teenagers in the United States aren’t outright banned from Instagram.

Meta’s longer-term ploy, which it disingenuously alluded to in its statement, is to drag down its competition. If Meta is meant to suffer, so too must TikTok and YouTube. More, from the statement:

“While this is an important step, the fact is that teens move fluidly between dozens of apps a day. All platforms should empower parents and support teens by putting the same measures in place, because we know that when teens are restricted on one app, they simply move to another. For meaningful progress to happen, we urge TikTok and YouTube to join us and state attorneys general in adopting this new standard.”

C.J. Mahoney, Meta’s chief legal officer, laid it on even thicker: “As a parent, I’m proud of both the work Meta has done to protect kids historically, and of this new groundbreaking agreement. But its success depends on all other social media platforms following Meta’s lead.”

As “the son of a parent,” let me just say: asserting that other platforms should “follow Meta’s lead” is an unbelievable heat check. Meta was sued by basically the entire country for a reason! And Meta agreed to a settlement so that it would not risk existential restrictions!

It’s true that if YouTube and TikTok agree to similar terms, the settlement contains several contingencies to tighten Meta’s social media regulations. Daily time limits could be reduced to one hour per platform, for instance. Meta would pay out more money, too. It’s a clever carveout by both sides of the settlement. State attorneys general can continue pressuring other platforms that have demonstrated a disregard for their younger users, and Meta gets to put the squeeze on its rivals. As Bloomberg’s Dave Lee points out, the average U.S. teen spends an hour and a half per day on TikTok and 34 minutes per day on Instagram. If U.S. teens were ultimately limited to one hour per day on each platform, that might even work out in Meta’s favor.

I have no idea what the other major tech companies with social media platforms will do next; I certainly don’t hold them in higher regard than Meta. I do know that Meta is not to be trusted, as reinforced by its immediate and desperate effort to control the narrative around this landmark settlement. Meta is like a beat-up boxer who narrowly avoided a knockout punch, then declared himself champion of the world. These are distracting and fruitless antics. We all know what we saw in the ring—and the judges’ scores are due soon.

Here’s what else we’re reading this week:

• Leopold Aschenbrenner, the wannabe-wunderkind behind the failed AI hedge fund Situational Awareness, recently married Avital Balwit, the chief of staff to Anthropic CEO Dario Amodei. Their wedding was reportedly officiated by Ross Douthat, longtime conservative columnist for the The New York Times, who just took a job as a correspondent at 60 Minutes. Something something Mad Libs joke, we live in a simulation, this wouldn’t make it past the writer’s room, etc.

• Billionaire investor Stanley Druckenmiller submitted an op-ed to the Wall Street Journal that was clearly generated with AI. The assigned editor(s) didn’t recognize obvious AI markers, and the WSJ published the op-ed. Druckenmiller then unapologetically acknowledged that he used AI to write the op-ed. This would normally be the moment where a managing editor steps in, issues a mea culpa, and updates the article with a disclaimer. Instead, the WSJ’s Editorial Page Director, Paul Gigot, has haughtily doubled down. He is happy to accept AI-generated writing. Here’s what Gigot sent Semafor:

  • “AI is a fact of modern life. People will use it to assist in their work and their writing, including with research, checking grammar, editing and more. The question for us is whether what we publish from contributors reflects an author’s original argument, and if the author has the standing and credibility to make it. In Stan Druckenmiller’s case, we have had a relationship with him for many years, and nobody can doubt that his op-ed is his genuine opinion.”

  • I can absolutely doubt whether Druckenmiller’s piece is his genuine opinion! He didn’t write it in full; we have no idea which thoughts are his, and which were spit out by an LLM. Editors at major news outlets should not be defending contributors who are too lazy to do the bare-minimum work required to share their perspective.

• Sam Altman is once again giving wishy-washy predictions about when OpenAI will achieve artificial general intelligence, better known as AGI. In a new interview, Altman told Time that his company is “not quite yet” there, but will have an internal system by year’s end that he, personally, would characterize as AGI. Not to be outdone, his colleague and cofounder Greg Brockman told Time, “We are looking at transforming the entire economy.” Maybe OpenAI could figure out if, or when, it’s going to IPO first.

• Google cofounder Sergey Brin is helping to fund an advertising campaign against California’s Prop 40, a ballot measure that would require billionaires in the state to pay a one-time 5% tax on their net worths. The first anti-Prop 40 ad notes that outgoing Governor Gavin Newsom, as well as the two remaining gubernatorial candidates, Xavier Becerra and Steve Hilton, oppose Prop 40. That is true, unsurprising, and an indictment of all three politicians. The ad also mentions a study from the Hoover Institute, a conservative think tank, which predicts doom-and-gloom if billionaires sacrifice a nickel. Notably missing from the ad: the word “billionaires.”

You may have seen headlines about Port Washington if you’ve been following the fight against data centers; friends and colleagues of mine with no connection to Wisconsin have heard about arrests at city council meetings, an attempted recall, and the landslide passing of a tax benefit referendum drafted in response to the data center.

Port Washington, Wisconsin is beautiful. On the shore of Lake Michigan, half an hour north of Milwaukee, a fishing town with a historic marina, and in the summers, it’s lush, vibrant, and a major hub for bird watching. It’s been compared to small town New England.

I drove right along the under-construction data center, all three-square-miles of it, on my way to meet the neighbors and organizers. It’s contained in blue privacy fencing with “Vantage Data Centers” branded every so often. I was meeting some members of Great Lakes Neighbors United, a group started by local residents that’s been one of the major voices of opposition to its development.

The neighborhood really is right next to the site. I walked up to the door of the address I’d been given via email; there was a note taped to it saying to walk right in and leave your shoes on, with a smiley face. The whole kitchen counter was a buffet. A couple of UW-Milwaukee journalism students were already chatting with some of the neighbors at the kitchen table. They all smiled, welcomed me, and encouraged me to take some food.

I found Amanda — the homeowner, I eventually learned — sitting on a couch on the back porch overlooking a lush garden. We immediately started chatting about our shared love of hiking and the beauty of Wisconsin’s nature in the summertime. It’s the reason she chose to move and buy a home here in retirement, all the way from San Antonio.

Conversations edited for length and clarity.

Amanda: We always thought this was a special place that time just forgot about. And I’m just so afraid. These people are just not prepared for what’s going to happen. And it’s unfair — they didn’t ask for it. They didn’t. This is nothing they had any say over. It was literally shoved down their throats.

We would go and we would yell at the mayor. I kept on saying, ‘Slow down, slow down. Take your time. You’re not thinking this through.’ And the reason I can say this is because I’m from San Antonio, Texas, and I lived next to three data centers for the last ten years. So I knew what was coming, and I knew how they did it.

That’s what was so hard about watching this come in. And seeing these people, a lot of these people have been here all their lives. It’s all they’ve ever known. And to see what’s going to be the change in culture, the change in everything. They don’t know what’s coming.

Tracy and Jeff joined us on the porch couch next. Tracy is a mother and mental health advocate whose family has been in Port Washington for three generations. Jeff moved to Port from Milwaukee in retirement; he shows up with a “No Data Center” button pinned to his ballcap. Both live in town, a couple miles from the data center, but have family in close proximity to the site. They, and Amanda, are members of Great Lakes Neighbors United.

Tracy: [GNLU] was started last year here. It’s basically a group to help those around the data center. I mean, it first started here, and now we’re branched out to help other data centers, other people who don’t know what to do about a data center coming to their town. We’re just trying to help spread the word of what could happen.

Native Wisconsinite Charlie Berens, a well-known journalist and comedian who has been leading and assisting with anti-data-center advocacy around the state, wore a Green Bay Packers tie in a video about Port Washington approving the massive data center in August of 2025. It now has over two and a half million views, and was how some Port residents learned what was imminent.

Share

Tracy: So last year, after I heard of the Charlie Berens video, I left the council meeting when they were still doing the public speeches. And I was walking home out of frustration. I’m just thinking, what are we gonna do? This could be detrimental for the environment. I had lots of questions going through my mind trying to think, what happens if this? What happens if that?

So I went on Facebook to see, is there any group that’s talking about what to do about a data center, saying no to this data center that’s coming to Port Washington or anywhere? And I couldn’t find anything.

So I created a group, shared it saying, ‘Hey, for those who went to the council meeting, I created this so that we can share helpful information on what to do, what not to do.’ And then from there, I got a few people, and I asked, ‘Hey, can I please get some help with monitoring the group?’ Because I have a full time job taking care of my kids. And from there a few people volunteered to help. One of them was actually a school classmate, and she’s the one that started Great Lakes Neighbors United.

In a state and locale that places so much value on nature and the environment, residents were appalled to learn no environmental impact report had been conducted. The state’s Department of Natural Resources has faced existential budget cuts in the last two decades — devastating for a state with a storied history of environmentalism.

Jeff: That’s the whole problem here to me, is the [DNR] cuts. Tony is taking care of the whole Ozaukee County.

JJ: One guy?

Jeff: One guy. That’s Tony. Tony’s the man. And if that’s happening here though, I mean, imagine some other areas.

JJ: Wisconsin Rapids can’t get ahold of a representative. They just get sent back information sheets.

Jeff: There you go. And probably because their rep is, who knows, in Iron county or something, 150 miles away from them. So. I guess we’re lucky we got Tony.

Cuts to the DNR made it easier for an environmental impact report to be overlooked. The group found out via Freedom of Information Act requests, however, that Vantage was directly involved in the omission.

Amanda: Come to find out — we did a FOIA on them, we actually got records on it — Vantage was really upset. They’re like, ‘If we have to do an environmental impact study, this project will not go forward.’ And the mayor said, and I can show you the emails, the mayor said, ‘Let me make some calls.’ And two weeks later, they’re like, ‘Clear.’

No environmental impact study was done. He went to Governor Evers. Governor Evers greenlit it. Which is a shame. I wish I could suck back that vote that I voted for him because he was always Mr. Conservation.

Tracy: There was something that Vantage needed them to look over. I think it was the generators, the emergency backup generators that they were supposed to approve of. But they basically let it slide because, if they approved it, it would have killed the project.

Jeff: Yeah. They didn’t do the environmental impact statement. I have concerns about the contaminants. And maybe not me, but are we going to have the worry in 15 years for — I have grandkids a mile away from it.

Amanda: We’re on a bluff. I mean, you’re 80 feet above the lake right now. Bluffs are notoriously unstable. They have a tendency to slide down and take houses with them. There’s an association of engineers that assesses your bluff for integrity. That was never done. You know, I mean, your bluff should have integrity. I told Heidi, when all the vibration was going — She lives on top of the bluff. And I’m like, ‘I need you to go to bed at night with a life jacket and a crash helmet, because you’re gonna end up in the lake,’ you know. That’s the worst thing you can do to a bluff. And not one study was done.

Subscribe now

Heidi is another neighbor in the subdivision, and also has deep Wisconsin roots. She and Amanda have kept a close eye on the site as Vantage moved in. It’s hard not to, frankly. There’s one road in and out, and it goes past the data center.

Heidi: I don’t know if you drove around it yet or up to it, but you’ll see blue plastic that goes around it. It says ‘Vantage.’ That thing is three miles square. That’s huge. They put up the, you know, the plastic, and they started just raping the land.

Amanda was driving her car down, and she saw a pile of dead deer. And they must have just— They didn’t care. We’re all big conservationists, you know, and environmentalists. They must not have cared anything about the wildlife within the walls of that fencing that they put up. And I’m sure those deer were so terrified when they started all this heavy equipment. They’re probably running in there, breaking their necks, breaking their legs. They had stacked up dead deer. And she was going to take a picture, and this security guard caught her. He came, and goes, ‘Put that phone down.’ She said, ‘I can take a picture.’ Is she a fighter or what?

Yeah. She said, ‘This is a city street. I can stand in the street and take pictures.’ He said, ‘I’m gonna take your phone and smash it.’ So that’s the kind of thing that was from day one. That’s the kind of thing we put up with.

It doesn’t stop them from taking any opportunity to walk or ride bikes along those roads. They jokingly call it their “spy ride.” If they notice any updates or new information, they let each other know. They certainly won’t be intimidated.

Amanda: It was like dawn, it was a Sunday. And I’m like, I’m gonna see. So I took my lawnmower, and I have a pretty beefy lawnmower. And it’s got a flag that says ‘bite me’ on it. I’m like, you know what, I’m gonna see what the reaction time is. I’m gonna see what would happen. So I went up to the road, and I waited until there was no traffic whatsoever, and I just gunned it. Two and a half minutes. I had four of them on me, asking me what I was doing. And I said, ‘I’m gonna mow.’ And he goes, ‘where?’ I’m like, ‘I don’t have to tell you where I’m mowing. I’m just mowing.’ And he goes, ‘Give me some ID.’ I’m like, ‘I’m on a mower. Where am I gonna put an ID?’

It should be noted: this was Vantage’s private security, not local police. She did not legally have to show them ID. It’s unclear if they were legally allowed to ask her for one.

Christine, who had invited me to speak with them, joined us in the yard next. She was born and raised in neighboring Washington County, and is almost finished earning her PhD in social anthropology.

I’d seen the disturbing videos of arrests at Port’s city council meeting the year prior. Christine was one of them. While recounting it, she and the others kept laughing in disbelief.

Christine: When I was arrested and dragged out of the common council meeting, the mayor was then interviewed because they were asking him about it. He said, like, ‘Well, you know, she’s an outside agitator.’ And then they said, ‘The women who were arrested? So they didn’t come from the city of Port Washington?’ And he had to say yes.

Tracy: They all came from the city of Port Washington.

JJ: So that was you!?

They laughed again.

Christine: Oh yeah.

Tracy: You’re famous.

JJ: Were you charged with something?

Christine: Disorderly conduct. Yes. They appointed a special prosecutor, first of all. And then they tried to give us a deal, saying, ‘Oh, we’ll knock it down to loud and unruly noise.’ And it’s like a discount on the fine of, like, 50 bucks.

More laughter.

Tracy: She said ‘Recall’ three times as she was going back to her seat.

Christine: And people were clapping.

Tracy: People were clapping. So it was just her voice over the clapping and stuff.

Christine: And like I said, not unusual compared to any of the behavior in the past.

Tracy: What was unusual was the officer, the. The sergeant at the time — now he’s retired — he quietly went to Christine to say, ‘I need you to leave’. And it’s like, you’re doing this as someone else is now speaking.

Amanda: If anybody should have been arrested, it should have been me and that neighbor, because we attacked Teddy [the mayor, Ted Neitzke IV]. We were screaming at Teddy, and it was— It was really fun. It was really kind of fun.

Leave a comment

They had a lot to say about the mayor. In several different conversations, longtime residents told me it was common knowledge that he’d been a bully in high school. He won because he ran unopposed. They said he’s been frequently taunting and condescending to residents opposing the data center. They referenced reporting I’d been following myself this month, Jasmine Sun’s data center work in the midwest, during which she met and drove around with Port Washington’s mayor. She described how he compared himself to Theodore Roosevelt, blamed Charlie Berens, journalists, and “outside agitators” for the opposition, and how he proudly told her there’s nothing he would have done differently in communicating with residents about the center.

Tracy brought up that quote from the mayor, to a chorus on the porch of “How cold can you be?” and “Heart wrenching.”

Tracy: Yes, I saw that. And I used that against the mayor at the last council meeting because it’s like, how could you be so heartless to not even take into consideration what’s going on? Like, I’ve heard so many people complaining about how this is destroying their lives.

Advocating against the data center — even just advocating for more information and transparency regarding it — is affecting their lives, too. One member has had cupcakes thrown at her house. A member of the GNLU who is a naturalized US citizen has recently begun receiving odd paperwork from the Census demanding an immediate response and threatening vague consequences if not. Several members have received odd anonymous letters.

Tracy: I won’t even go into details about this letter, but it was so morbid that I took it to the police. And I looked up the address of where it came from. It’s an abandoned building from Illinois. And I’m like, what in the world’s going on? That’s why I put up the security camera, because I fear for my children’s safety. I fear for my safety. And it’s never been like that for Port. It’s like, you could just keep your door open. No one’s gonna barge in. You could keep your doors unlocked for your cars. Now? No.

There have been other people within Great Lakes Neighbors United that have spoken out, that have gotten some iffy letters and they’re not signed from a person or organization, it’s basically from the same address that I got. Yeah, we don’t know who it is.

I mean, I’m not gonna stop what I think or what I’m for. But those fears are out there nowadays. Definitely.

They’re also highly aware of how this data center could be used. OpenAI and Oracle intend to move in once it’s built. Nobody can give them clear answers on what will be stored, produced, or developed there, but neighbors notice the intense security and surveillance already.

Amanda: AI in itself, I mean, without having your community destroyed, just the fear of what are they gonna do with this. When you came here and you crossed that road, I don’t know if you noticed or not, but you went through seven Flock cameras. When you leave, you’ll see them. There’s seven Flock cameras on the corner. There’s only one way out of this subdivision. So every time we leave, they know exactly what our patterns are. They know. If a friend of mine comes over and gets in my car, the Flock camera reads your license plate, and they can ping their phone to find out who I’m riding with, just from their cell phone.

I mean, this is the way surveillance is. And we’re in the baby stages of it. Why do they need so much surveillance? Why do they need to know everything that we’re doing, every place that I’m going, what my schedule is?

Yeah, I have a lot of questions, and I’m so afraid that I wasn’t really concerned until Anthropic was dropped and they picked up Oracle, that they’re going to militarize Oracle. This is the largest data center going. They’re really investing a lot into this one, especially with security. Is this going to be another military installation? Is this going to be there — are we building the end of the world?

Moving away isn’t an option. Residents didn’t know it was coming until it was too late. Their property values have plummeted. They’re trapped.

Amanda: I had a realtor come last week to talk to me, and he pretty much said if I want to sell the house, I have to [lose] 150,000 to 300,000 because nobody will live next to a data center. No one. So we lost that much value on our house already. I tried to sell it last year before shovels even went in the ground. Then as soon as I disclosed it with the data center, I had two full price offers torn up right in front of me.

I cannot sell this house. And what’s worse is the realtor told me not only can I not sell this house, but I also can’t go look at other houses because it’s contingency. And once they see the address and then I’m so close to the data center, they’re like, we’re not taking that. So I can’t even look at other houses.

Share

Jeff: I can tell you, with my son being that close to it, he kept saying something was going on, but for a year ahead of time, nothing was brought out with what was actually going on. When the signing came, we were at all the monthly meetings.

I’m still against it. I’m still affected by the constant traffic today. You can drive through and you can see the dust in the air, and that’s minimal of what’s been going on. And that’s not even what’s coming.

Water contamination is a massive concern. In the neighborhood next to the data center, every house has its own well. Even more urgently, a walking distance away is Lake Michigan.

Currently, more than 30 million people get their drinking water from the Great Lakes. The five lakes are one massive interconnected watershed. They account for over 20% of the Earth’s accessible fresh water, and over 80% of North America’s. They are an indispensable ecosystem to the health and future of millions. The stakes of the health of that water are indescribable.

Vantage is already using Port Washington’s water in ways that are legally dubious. The neighborhood next to it is most immediately threatened. With no environmental impact report, there is not adequate information on the effect it will have on the Great Lakes.

Jeff: I’ll call it the far fetched hope that we could still stop the project. Hopefully they can’t get water, because now they want to use well water. And I mean, to me, that’s crazy to begin with, but crazier yet is that they’ve allowed them to use these wells all the way up to this point.

If that were you or I demolishing something to say, ‘We’re putting up a barn here,’ that well would have been capped and covered and sealed. Yep. No, not for Vantage. Vantage can.

Jeff: We’re battling with these wells right now. They need the water first, you know, or the approval, and they keep using it when they’re not supposed to.

Tracy: They have the water.

Amanda: They’ve been using it all along.

JJ: So what happens, they just get fined?

Jeff: No, you don’t hear of fines.

Tracy: They won’t get fined.

Jeff: Are you going to start having contaminants because you’re sucking the bottom of a well? I know when the water goes down, there is none.

There’s also the water usage that’s being used by a power plant, because they intake Lake Michigan’s water and run it through and use that for cooling. Now you heat up the bay, now it evaporates more. And that’s just a simple concept that I’m grasping. There’s other ways that it’s going down while they’re using it. So I mean, it’s not just the water they’re using there, it’s the water they’re also going to be using at the power plant.

Heidi: If you have never seen that, you don’t realize. It’s spewing out warm water. There’s warm water going out into Lake Michigan. We’re not the only city that does that. We’re just one of many. You know, Milwaukee, over in Michigan they’re doing it. So it’s not a small deal. It’s a huge deal.

Amanda: And then you’ve got ambient heat because a data center that big is gonna change the air temperature ambiently around it. It could be, it could be 2 to 3 degrees anywhere to 16 degrees. And that’ll change the whole ecosystem.

Port Washington is already a unique microclimate because of the atmospheric effects of the Great Lakes system. Residents dryly call the area the “Cancer Crescent.” Once again, they have no answers for how the data center, its water usage, or the inevitable new power sources required to power it will affect the safety of the air and atmosphere.

Christine: The air quality is a huge question that people have.

Tracy: Ozaukee county already has bad air quality as is.

Christine: Yeah. It does not meet the federal standards for ozone.

Amanda: You know, maritime winds, they always prevail from the north. And then most of our winds always come from the west. If you watch the weather at all in the summertime from Milwaukee to Sheboygan, they’ll always be ozone alerts for the area. We call it the Cancer Crescent, which is, you know, a bad name, but it’s because the ozone always pools. They built a data center right in the middle of where this air does this. So not only do we have air that’s not going to dissipate, it’s going to literally stagnate right over us. So whatever pollution we get from that thing, if they do have to run gas turbines, which is what they’re threatening to do, it’s amplified. It’s going to be. I mean, there is no control. We did not have an environmental impact study done.

Christine: Actually yesterday marked a year, August 19th, when the developers’ agreement was signed. It was Aug. 19, 2025, and a couple of us went to the common council meeting. It was full of people from the surrounding counties who were finding out just then that they were going to be affected by power lines coming through their properties. And that was the same day that our developers’ agreement was signed.

So, like, all of our attention in the city of Port Washington was on the data center site itself, you know, where we were focused on that. But then it’s like, well, wait, yeah, how is this thing going to be powered?

Now it’s looking like the transmission lines are becoming increasingly delayed. But they don’t even know if they have transmission lines ready. We don’t actually know where they would plug into, like, the energy source. You’ve got this transmission line to nowhere, basically. That’s why we’re really concerned about on-site power generation being an option, because that’s what Vantage has used. In Sterling, Virginia, they found a loophole with which they could run generators 24/7. And those were diesel generators. But we’re also very concerned about the methane gas turbines. Like what’s happened in Memphis with Elon Musk.

The data center isn’t in the city of Port Washington itself, but in the township slightly to the north. That’s made avenues for official opposition even more complicated.

Amanda: We actually are outside agitators. We live the closest to it and we’re, we don’t live in the city of Port. We live in the town of Port. And that data center took up two thirds of our town. So we’re the little sliver that’s left. We’re kind of gerrymandered, so we don’t get any tax breaks. We don’t get any tax incentives. We can’t vote. We can’t vote against the mayor, we can’t vote for the mayor, we can’t do anything. And we’re suffering the worst from it.

And my biggest thing is, we’re an experiment. We’re guinea pigs. It’s like, can you live with this thing?

Christine: My sister ran into the guy who was heading up security when they first started, like in January. Her friend got to talking with this guy, and he’s like, ‘Well, you know, locals. They always put up a fight in the beginning, but then, you know, they just get used to it. They’ll quit their complaining.’

So that’s the mentality. ‘These locals who are in the way.’ Yeah, we’re in the way.

Everyone, from those who moved to the area in the last decade to those whose families have lived there for generations, told me the process of bringing in Vantage has changed Port Washington’s culture for the colder and more impersonal. It’s a good time to insert that on the homepage of Vantage’s website, they tout their slogan: “Where innovation, sustainability and respect for communities are built in from the start.”

Tracy: Growing up, it was a quaint little town, city. Very pleasant, very, very friendly. Like your Southern small town where everyone’s waving to you. And now with the data center in, it feels like there’s so much tension that it’s losing it. Because of all the stress that the data center’s brought, it’s changing the culture. It’s really changing the culture. It’s like it’s lost its charm. It’s lost. It’s like almost friendliness, almost.

Right now, I’m advocating for mental health because I’ve seen people getting stressed, people losing sleep, and when I’m talking to them, I let them just let their frustration go because it’s like, don’t bottle it in right now to the point that someone breaks around here. I mean, it could be a crisis for their family, or it could be a crisis, let’s say they do something stupid within the city at some event, school, whatever, there could be a shooting, there could be a death. There’s that kind of fear of what could happen if mental health is not taken into consideration. So that’s why for the past couple months I’ve been going to the board meetings to advocate for the city to provide mental health to residents. Because I can tell you mental health is not cheap to get.

Jeff: Animosity throughout the entire city of Port. When I was canvassing, we all got along, we had block parties where we loved one another. Now, you can’t talk to half of the people and they don’t talk to you because of this issue.

Cassie Steiner, a senior campaign coordinator with the Sierra Club’s Wisconsin Chapter, told me they are already seeing more and more American climate refugees, and that many from regions affected by droughts, hurricanes, tsunamis, and tornadoes are already looking to the Great Lakes region as a climate safe haven.

The incentivization and deregulation of data centers in the region will have a detrimental effect on its environment, incomparable to its previous industrialization. The scale of electricity data centers use is magnitudes larger than a standard factory or mill.

Steiner told me a recent study warned the water in the Great Lakes is not renewable in the way water resources usually are. “If there’s enough water withdrawn from that watershed, and it’s being put back elsewhere, or evaporated up and raining down elsewhere, that changes the characteristics of the water, how much there is, what level of pollution there might be,” she said. “I think we take clean water for granted at some points. It’s a huge concern here. And that water is part of why data centers are looking at the Midwest.”

Amanda: I’m really disappointed in Wisconsin. I just didn’t. I never thought they would do this because they were such stewards of the land and they were, they took care of the land. I mean, this land fed them. I mean, they understood how it worked.

I’m a big hiker. One of the things that I always noticed about Wisconsin was that they don’t litter. There was never any litter. I mean, I could be like 50 miles, you know, in, and there’s never, not even a cigarette butt. These people were good stewards, you know, and to see this just ripped away from them for an unknown entity that we’re not even sure what it’s for or what it’s going to do, it’s hard.

Christine: For those of us who moved here specifically because of the environment, we didn’t move here because we love a data center. To the contrary. Had we known that this would be coming, I never would have considered — I mean, I grew up around here, so I know of plenty of other places that would have been great. I invested the past five years in this community, and living here, and thinking this will be a place to put down roots.

I constantly feel in my life now, this is the second time we’ve, my partner and I, have owned a home. The first time being in Florida, when we were flooded out because of overdevelopment due to impervious surfaces that pushed water down. We sold and left because of the climate issue there. And I would very much consider myself a climate refugee.

Coming back to Wisconsin, you know, I was like, yep, I grew up there. Don’t worry. Things move slowly. People are considerate, and deliberative, you know.

And then this. Are you kidding me? Like, it’s just like you, Amanda, you left Texas. You were being surrounded by data centers. And it’s like you come up here. Again, it’s like, are you kidding me? My adult life is constantly being uprooted by a natural disaster due to climate issues, you know, overdevelopment, lack of planning. It’s becoming really exhausting.

Amanda: ‘Unfortunate’ has become the word that triggers us because everything — ‘It’s really unfortunate that we couldn’t buy your house because you live next to a data center.’ ‘It’s unfortunate that we have to lose a hundred thousand or two or three hundred thousand if you want to leave.’ Unfortunate is always the first word they use.

And it’s, it triggers me. I said, you know, it’s not unfortunate that we chose to live here and have land and houses that we knew what we were buying. All of a sudden to be ripped away and our future is gone, and we don’t know what we have anymore, because it was stolen from us. Our future has been stolen.

Though the data center is under construction, and some residents resigned to it, the group told me they aren’t giving up yet. The Sierra Club has an outstanding lawsuit against Vantage.

Christine: I think you can’t allow them to colonize your mind with their mantras that this is a done deal. Otherwise, you’ve already lost and there’s no point.

I spent the better part of six hours at Amanda’s house. She, and everyone I spoke with, were instinctively warm, even through anger and devastation. There was continuous laughter; they told me they had to be able to laugh. They made Terminator references, joking that some saw them as the real-life Sarah Connor. A five-month-old floppy puppy named Sunshine romped into the yard from next door, looking for Amanda’s dog to play with. A sweet elderly couple who live on the street stopped in to say hello on their way to play cards in town. I have no doubt the culture of Port Washington has changed — the footage of residents being dragged from city council meetings really is sobering — but the culture they described as indispensable still prevails here, not even a half mile from the Vantage project.

And only as I was getting ready to leave, holding the sandwiches she’d insisted I take and with an invitation to return, did Amanda turn to me and ask, “By the way, who are you?”

Part of what’s so harrowing about all of this is that typically, if you are being digitally stalked or harassed by an ex-partner, your main recourse is to report it to the police. Here, not only were local police the stalkers—they were given free rein to indulge in their stalking tendencies, via the surveillance infrastructure their departments spent millions of dollars on.

While some are quick to vilify the police officers as bad men tantalized by a new toy, Flock’s tentacles have been spreading for some time now. And it is impossible to detach Flock from a broader ecosystem of control through technological surveillance.

I first heard about Flock relatively recently, at a conference in April organized to “take back tech” from oligarchs. At this conference, community organizers discussed Flock’s sprawling architecture, how their automatic license plate readers (ALPRs) capture images of cars and license plates virtually everywhere—tracing the whereabouts of cars not only through mounted cameras but also through drones and mobile trailers that do not need permanent mounting, fixed utility poles, or wired power.

These community activists got involved in this fight long ago because Flock, for years, has been tracking vulnerable people via their license plates—from immigrants seeking refuge to women crossing state lines to get abortions.

In the name of public safety and “eliminating crime,” as Flock laughably puts it, Flock cozied up to city council members, inking multimillion dollar contracts between local police departments and its surveillance infrastructure.

Local activists were only able to dismantle these agreements by understanding how deeply unpopular these cameras are across many different political affiliations. What was meant to be a public safety measure was actually making people feel like their every move was being watched. So activists built coalitions, pressuring any council member who supported Flock with the loss of their seat. Eventually, in places ranging from New England to Texas, voters made it very difficult for pro-Flock council members to retain their seats.

As it turns out, it is a nearly universal human experience to rather despise being watched without our consent. The freedom to move around is also a core tenet of the United States. In fact, the freedom to move around is probably how every single car manufacturer listed in Flock’s database has marketed itself for decades.

And now, unlike the invisible surveillance infrastructure humming in the background because of tech firms like Palantir, we can physically see Flock cameras watching us—perhaps more graspable as an invasion of privacy in an emotional sense, proven by the people who literally sawed down a Flock camera in Kansas City the other day.

All of this has allowed progressive midterm candidates to leverage right-leaning disdain for Flock, banking on an anti-Flock platform to swing libertarian voters in their favor. For instance: Abdul El-Sayed, the Democratic nominee for Senate in Michigan, has made one of his key campaign talking points the fact that his Republican opponent is tied to Flock cameras. Representatives in Ohio and Tennessee are also calling out their opponents’ affiliations with Flock. Even President Trump was asked about Flock at a press conference last week.

Meanwhile, responses from Flock and its backers are feckless. In the Washington Post article about the stalker-policemen, an Andreessen Horowitz partner who invested in Flock tried to deflect from the police officers’ malicious usage, arguing that the number of times Flock cameras have been used improperly are “infinitesimally” less than the times Flock cameras have been used properly or for positive law enforcement reasons. In a totally bizarre follow-up, he compliments Flock for rooting out bad apples in police departments, claiming very brazenly that we should thank Flock for identifying the stalkers in their departments.

Of course humans have problematic and bad tendencies. (This is why Flock will never “eliminate crime,” unfortunately.) But there is literally no way for a Policeman Ex-Boyfriend to stalk his citizen ex-girlfriend hundreds of times a week without something like Flock—unless he were literally tailing her physically, in which case he probably would be fired for not showing up to work. Beyond that, this investor’s defense is the same tired argument used for everything from guns to social media: that the tool is not the problem, but the person behind it. Yet time and time again— because of incentives that must only benefit the shareholder like this Andreessen investor at the end of the day—there is an emphasis placed on the individual, and an unwillingness to acknowledge any role that technology or systems have in this equation.

Here, finally, it’s impossible to ignore the groundswell of opposition to Flock, and the swift action that’s been taken in the last month. Even police departments are holding their officers accountable: in the last week, four Georgia police officers were arrested for using Flock to stalk. A constable in Texas told Flock to “come get your garbage out of my county.” On a milder note, it’s been proposed that surveillance camera companies must institute privacy guardrails within the product to protect from abuse—like requiring police officers to label every license plate search with a criminal case number.

But at the end of the day, this is about something more than just Flock. As people learn more about this Black Mirror plot unfolding in real life, Flock has risen from relative obscurity to something not only immoral, but something terrifying. Most people don’t love the idea of being recorded on their bathing-suit-clad way to a typically safe public pool. They certainly don’t like the idea of their local deputy microwaving buttered popcorn between midnight and 4 a.m. to watch his ex-girlfriend’s new boyfriend drive to the pharmacy.

And when technology becomes terrifying, that may overshadow the purported “convenience” or “public safety” of it all. People will start to feel harassed, unsettled, controlled. And you know who really hates feeling like they are controlled—or like they have no freedom to drive their car? That may be a very large swath of American voters.

Share

What else we’re paying attention to…

Australia is banning generative AI from topping the music charts.

The AI-focused hedge fund Situational Awareness received subpoenas from the Securities and Exchange Commission, asking for its communications with lenders about the money it was borrowing.

For the first time, New York City has more tech workers than San Francisco, attributed in large part to Wall Street’s focus on AI.

Boeing engineers and technical workers—part of a 17,000 person union, the largest white-collar union in Boeing—voted to reject the company’s four-year contract offer.

Despite owning property there, Jeff Bezos and Jared Kushner can’t get into this one country club.

Jon Oliver digs into predatory subscription practices, and corporations that engage in “dark patterns” to keep you locked in and spending more. “No company needs to know I had a child. There is a reason that you start birth announcements with dear family and friends, and not, ‘To whom it may concern at the Whirlpool corporation.’”

A popular AI assistant has embedded itself into email, messaging apps, calendar, and their users’ device’s audio, location, and screen. Some users—citing times when the AI agent became “naughty” and sent out emails without consent—are concerned about privacy.

An arbitrator ruled that the Washington Post improperly fired union-protected columnist Karen Attiah for her posts about Charlie Kirk.

Potentially losing your 27,000-follower channel while promoting your book would be far from ideal. Durán didn’t consider the timing to be a coincidence. A Bay Area journalist, he’s been tangling with tech oligarchs and some of their most sycophantic believers for years, including during the recalls of San Francisco school board members and the city’s progressive DA, Chesa Boudin, in 2022. Tech oligarchs are not fans of Durán; in April, his X account was permanently banned for responding to one of Palantir’s manifestos with “TLDR: Fascism.”

As Durán writes in “The Nerd Reich,” which was released on Aug. 18, we’re long past the days of (perceived) benevolent billionaires who simply hold the “eccentric beliefs of rich men who hated taxes.” In fact, he said, “this framing is now dangerously obsolete” and “has now curdled into something even more sinister: tech fascism.”

The Silicon Valley-backed San Francisco recalls were just the tip of the iceberg. Top tech figures have bigger aims, Durán said. Their big break came in 2024, when they helped secure another presidential term for Donald Trump and the MAGA movement. Trump’s crypto glad-handing, and the ascension of his vice presidential pick, J.D. Vance—a now-reticent protege of Peter Thiel—have made Durán’s definition of tech fascism all the more prescient.

“Tech fascism bundles powerful technologies—AI, crypto, defense tech, social media platforms, surveillance systems—toward authoritarian goals,” Duran writes. “These technologies are bound together under the control of a small, interconnected group of oligarchs who, simultaneously integrated into the highest levels of government, are creating an authoritarian infrastructure with boundless power. Its goal, like MAGA’s, is to privatize the public and eliminate democracy.”

Among the powerful characters profiled in “The Nerd Reich” are Thiel, the pseudo-intellectual who funds tech fascism in the shadows; Curtis Yarvin, a far-right blogger who’s fervently anti-democracy while admitting that he’s “not exactly allergic” to white nationalism; Vice President Vance, who’s a fan of both Thiel and Yarvin; Balaji Srinivasan, a wealthy venture capitalist and obsessive media-hater who penned a book called The Network State: How to Start a New Country; and Elon Musk, who’s ideologically aligned with the tenets of tech fascism, even if he’s not quite smart enough to be a spokesperson for it.

“The Nerd Reich” is sobering and terrifying. It pieces together the tech fascist movement in America, as well as ongoing attempts at establishing anti-democratic “freedom cities” around the world that are ruled by wealthy right-wingers. And it makes a convincing case that Thiel and Srinivasan in particular are existential threats to an already-fragile geopolitical order.

I admit one of Thiel’s pet projects does not intimidate me, though. I’m referring to “seasteading,” where you set up an autonomous settlement in the middle of the ocean. Thiel reportedly gave $500,000 to help launch the “Seasteading Institute,” a nonprofit that has seemingly struggled to stay afloat, if you will.

There’s other good news in the fight against tech fascism, Durán told me. “Billionaires have become very, very unpopular,” he said. “Even MAGA voters understand that the tech billionaires are bad. Tech billionaires are really not polling well now. And this coincides with a movement against data centers and a big attitude shift against AI, which people perceive as a danger and a threat to their families, to the economy, to their lives.”

Another positive sign: Durán’s YouTube channel isn’t going to be deleted after all (at least for now). The channel’s pending strikes have been resolved. Durán believes that whoever filed the strikes realized they were probably giving him even more publicity; YouTube didn’t respond on the record to my inquiry.

Excerpts from my recent conversation with Durán are below. They’re edited and condensed for clarity, and also presented as video clips. And if you’re interested, you can buy a copy of “The Nerd Reich” here.

On how “The Nerd Reich” came to be

On Peter Thiel’s enormous influence among tech oligarchs and the Trump Administration

Gil Durán: Peter Thiel is known as the godfather of the PayPal Mafia, a group of oligarchs who got their start at PayPal in the early 2000s. And he is also the godfather of the Dark Enlightenment or the Neo-Reactionary Movement, which is the name people have given to this extreme right-wing ideology that’s come out of Silicon Valley over the past decade. If you look at the people involved in these movements—Curtis Yarvin, the political guru who openly calls to replace democracy with dictatorship, and Balaji Srinivasan, the founder of the Network State cult—they say that democracies should be replaced with corporate-controlled territories. They’re all proteges of Thiel, who’s been able to create a cult of anti-democracy belief among influential people in Silicon Valley.

Thiel is a pseudo-intellectual. And nothing makes that more clear than the fact that much of his political worldview was inspired by The Sovereign Individual, a kooky apocalyptic conspiracy theory book that predicted different technologies would destroy democracy in the 21st century by undermining nation states. It predicted a cyber currency that would allow wealthy people to cross borders and hide their money from the government. Today, we would call this crypto. And the other technology was advanced automation, which we would call AI. As a result, countries like the United States would collapse, and it would be the dawn of a new era in which wealthy individuals freed from the chains of democracy and government would become so-called cognitive elites or sovereign individuals who would travel the world with multiple passports. They’d evade taxes, laws, and regulations, and live free of any society or country or nationality.

The book was very poorly reviewed at the time. People thought it was implausible and apocalyptic and unconvincing. But it had such a great impact on Thiel that he’s actually named it as part of the inspiration for PayPal. When the book was reissued in 2020, Thiel wrote the forward.

On Balaji Srinivasan and his attempts at establishing freedom cities

Gil Durán: In 2022, Balaji Srinivasan wrote a book called The Network State: How to Start a New Country, which laid out a blueprint for creating these new network societies that would eventually buy land, crowdfund territory, and create new nations. It’s basically the cult of The Sovereign Individual, backed by billions and billions of dollars.

In reality, building these corporate territories has been a lot harder than Balaji assumed. The best example to use is Próspera in Honduras, where a group of billionaire venture capitalists, including Marc Andreessen, Peter Thiel, et al, have supported a new sort of tech colony on the island of Roatán. They were able to carve out a special zone where they have different laws and rules. The people of Honduras are rejecting Próspera. In fact, the Supreme Court of Honduras ruled that it was unconstitutional for the previous government to award this sort of special zone. But now they’ve got a new president who’s very pro-billionaire.

In addition, you have this idea for something called Praxis, another billionaire-funded city that was supposed to be built somewhere in the Mediterranean, and then was going to be built in Greenland when Trump was threatening to take Greenland, and then was going to be at Vandenberg Space Force Base in California, but has never really found a home.

Most recently, Balaji Srinivasan created something in Malaysia called Network School, where he took over an empty apartment building and claimed it was the beginning of a new nation. He had a bunch of digital nomads go there to participate, but the Malaysian government caught wind of it. There was a whole drama about whether Srinivasan was allowing Israelis to attend, because Malaysia and Israel don’t have diplomatic relations. To make a long story short, Malaysia kicked out the Network School.

Balaji Srinivasan demanded a meeting with the prime minister and threatened that Silicon Valley would divest from Malaysia if he didn’t get it. He’s now announced plans to build a new Network School in Kazakhstan, which is an authoritarian country, so it’s perfect for these oligarch fascist types. The authoritarian president there wants to build a new crypto megacity in Kazakhstan.

There’s also actually a proposal to build a new kind of freedom city on the ruins of Gaza. Jared Kushner, Trump’s son-in-law, is involved. Trump posted an AI-generated video showing a Trump casino on the beach in Gaza, money falling from the sky, and Elon Musk sunning himself. So these ideas are clearly in the Trump sphere. While a lot of the projects have failed to find homes or been rejected by real countries, the thing about billionaires is they’re really dumb and they have enough money to be dumb for a very long time. They’re going to keep trying and they’re going to find people weak enough or corrupt enough to go along with their scheme.

On J.D. Vance, an acolyte of Peter Thiel

Gil Durán: J.D. Vance is in a real pickle. He owes his entire career to Peter Thiel, who made him a venture capitalist in Silicon Valley, funded his venture capital firm when he went back to Ohio, and dropped a historic amount of money into J.D.’s Senate campaign. And then he made peace between Vance and Trump, because Vance had previously called Trump “America’s Hitler.” But Vance has risen to power at a time when billionaires in Silicon Valley are unpopular with a wide swath of American voters. So now he’s trying to pivot into some kind of religious figure and some kind of humble guy who’s trying to be a Christian statesman, which I don’t think anybody who’s familiar with the things that J.D. Vance says or does would mistake him for a Christian or a statesman.

I read J.D. Vance’s recent memoir about his religious conversion, which he also owes to Peter Thiel. The ideas that inspired J.D. Vance to become a Catholic at the age of 35 were ideas he got from Thiel. He actually says that in the book, because he’s written about it so much in the past he couldn’t really hide it. But what’s missing from J.D. Vance’s memoir, which covers him going from being a law student to being the vice president of the United States, is Peter Thiel’s critical, essential role in every step of his career. That’s all erased from the book, which I found kind of hilarious because it shows you how afraid he is of having that connection, that dependency on Thiel exposed. And that’s why in my book I include an entire chapter on Vance’s relationship with Thiel: Maybe for the first time in history, we have a man one heartbeat away from the presidency who is completely owned by a radical, extremist, anti-American billionaire.

On Elon Musk and where he fits into the tech fascist movement

On crypto’s corrupting influence, and MAGA’s (current) stranglehold on tech fascism

Gil Durán: Trump has become a crypto billionaire since returning to office, and is part of the most openly corrupt administration we’ve ever seen. We’ve found out that crypto has a real use case besides drug dealing, human trafficking, money laundering, and terrorism: it’s a weapon of mass corruption. Suddenly, crypto has gone from not being a player to being the biggest player in American politics.

As long as they’re pumping money into the elections and backing up Trump with their influence, their algorithms, and their brands, the alliance between MAGA and tech oligarchs is going to remain a very powerful threat to democracy. It may not be perfect. They may personally not like each other too much, but they are united in a common project to destroy the pillars of democracy and make this country a fascist oligarchy.

On whether Democrats will stand up to tech oligarchs or ally with them

Gil Durán: The Democratic Party is very weak. And the only people more addicted to billionaire money than Republican politicians are Democratic politicians. I think people like Gavin Newsom are so hungry for billionaire money for their presidential campaigns that they are largely not attacking or criticizing the Silicon Valley fascist oligarchs directly. They’re going after Trump or after Vance, but Trump and Vance wouldn’t be able to do what they’re doing without the backing of the tech billionaires.

You have a project in California called California Forever, which is clearly very very related to the Network State idea. A bunch of billionaires secretly spent $900 million to buy up 60,000 acres of land in rural Solano County in Northern California to build a city in a place where there’s no water, no roads, no infrastructure, a whole list of endangered species in the area, and a law that specifically prevents a new city from being built there. The people of Solano County rejected this idea of building this new billionaire city. Now the billionaires have hired Governor Newsom’s consultants and they’re trying to push a secret deal through Sacramento that would force the city on Solano County.

It’s clear that 2028 is going to be a referendum on billionaire power in the United States. And it’s clear also that politicians like Gavin Newsom are servants of billionaires. And he’s not the only one. You’ve got Democrats voting for crypto bills. You’ve got Democrats talking about crypto as some kind of innovation tool. And that’s because they’re afraid. Crypto has made it clear that if you vote against their interests or even criticize them, they’ll spend to destroy you in a primary.

What we need are Democratic politicians who are not cowards, who are not afraid, and who have the moral gravity and the clarity to challenge this thing directly and courageously. We have to make it so that the more money crypto spins against you, the more people love you. We have to undermine the power of their money with something far greater than money, and do so with real moral courage and a bit of a revolutionary spirit. We have to crush billionaire power in the next few years, or billionaire power will crush us. In a democracy, people should not be afraid of billionaires. The billionaires should be afraid of the people.

In places like Wisconsin Rapids, they’re real and urgent.

Jasmine Sun has done some absolutely incredible reporting on data center backlash in Michigan and Wisconsin this summer. I followed the story less as a San Francisco-based tech reporter than as someone who has lived in both of those states, and loves the Great Lakes region deeply. As Sun points out, though Wisconsin isn’t one of the top states where data centers are appearing, it is an urgent problem here. And its culture, history, and status as a swing state make it a unique backdrop for a fight that’s only gaining prevalence nationally.

I grew up spending summers with family in Wisconsin; they’ve lived there for five generations. The beauty of this state inspired the birth of the environmentalist movement and Earth Day, and it’s where John Muir himself fell in love with nature. Sun’s description of it as Eden is earned.

I can’t claim to be from here, as much as I’d like to, but I do truly love this place. That’s why I wanted to know more about what this issue looks like in Wisconsin. I live in San Francisco now, where AI and the companies unleashing it seem to be conquering any aspect of culture they can reach as quickly as they can.

Wisconsin Rapids is a small city with about 18,000 people in the center of the state. Its history as an industrial hub predates Wisconsin’s statehood. The city straddles the Wisconsin River, where the state’s largest paper mill operated on its banks for over a century. The mill was the city’s main industry and employer until it closed in 2020. It was a devastating blow.

Now, the city is considering a proposal from the Russian-founded PNK Group to turn the closed mill into an AI data center. Residents are fighting hard to stop it before it can start.

On Sunday, I met three members of South Wood County Neighbors for Responsible Development for coffee in Wisconsin Rapids. I couldn’t drive more than a block through town without seeing a Stop the Data Center sign.

The three women I spoke with are not career activists. They each work full time. Kaitlyn Federwitz is a mom of two and the owner of a vintage shop. Emily Oetzman is originally from Lodi, in southern Wisconsin, and moved to Rapids for work. Danielle McDonald was born and raised in Rapids, and her family history here is the stuff of legend: her great-grandfather was the first mayor.

The first thing I asked about were the signs I’d seen driving in.

Conversation edited for length and clarity.

Kaitlyn: This gets into a whole story. So, there was a bar over, when you come into town, that was willing to distribute our signs for us. And we had a couple events there. People could come buy signs, and T-shirts, all the stuff.

Well, for completely personal reasons, [the owners] decided to close the bar. So we were then without a distributor and a place to set up. We started distributing signs at the Hotel Mead here, at their bar. The manager down there in the bar was like, yeah, we can, we’d tell people you can go pick up a sign here. It was easier than us delivering.

One of the Aldermen in town sent a threatening email to the Hotel Mead saying, ‘you shouldn’t be distributing their signs. 1716—that’s the [first] bar—they closed. And they closed because they were associated with this group. I just care for you as an Alderman and as a member of the Chamber of Commerce.’

She smiled.

Kaitlyn: So we’re recalling him. We filed the recall papers last Monday.

I asked if they’d been in contact with other groups of Wisconsin residents fighting data center development in their communities; cities on the southeastern lakeshore, like Port Washington and Kenosha, have had more media coverage of their opposition. A coalition called Great Lakes Neighbors United — who I’ll get to meet with tomorrow — has been doing great work in that region.

The Rapids coalition told me they had been in contact with GLNU, but came to understand that organizing in a place like Rapids has a different context than in the greater Milwaukee region.

Kaitlyn: We have had a few people who’ve offered to advise us, and I feel like we’ve kind of pursued it. But also, what we have going on here is a little different, because we’re not on farmland. It’s zoned industrial. The process that our city goes through, legally, feels different. A lot of other areas have the opportunity to change their zoning, all those kinds of things. Those avenues are not available to us.

Emily: Even if we change the ordinance, it has to be changed in such a specific way for us to prevent a data center from being there. But [the planning commission] voted to explore a moratorium at the last planning commission meeting after quite a bit of public pressure at the meeting.

Kaitlyn: You know what’s interesting, they were only supposed to discuss— I was waiting for that to be brought up. They were not supposed to vote on that day. The way it was written in the agenda was ‘discussion only.’

Emily: And then someone said that, at the mic. They’re like, ‘hey, our discussion is over. That’s what was on the agenda.’ And everyone in the crowd booed, and was like, ‘Just vote on it right now. Vote on it right now.’ The crowd went crazy until they sat up there and voted on it.

Public meetings and forums in Rapids lately have been packed and tense. One meeting was cancelled last minute because so many residents showed up, they were in violation of the building’s fire code. There are protests outside, screaming matches inside, and at least from those willing to speak, a united front against the data center.

Share

Kaitlyn: [The most recent meeting] did get tense. It got tense to the point of making me uncomfortable, because there were people who started shouting. Other than tense, I struggle to describe it. I almost thought someone was gonna have to stand up and say, ‘calm down. We can’t have a coup.’

Sometimes people associate anyone who’s anti-data-center with us, like we’re telling them what to say. We’re not—we’re a group of, like, four or five organizers.

Danielle: We can’t control what other people are gonna do. I just saw a comment on Facebook recently that we’re up in arms about things. That’s not what we’re trying to do. We’re not trying to get people up in arms. We’re trying to get people informed, and we’re trying to get them the information that they need. You know, because most people are saying, ‘I don’t know anything about a data center. I don’t know about this.’ So we’re giving them the information, because our leaders aren’t doing it.

The deliberate lack of answers, information, and transparency from their city government is a consistent frustration to them, and a throughline among Wisconsin organizers through each city’s different circumstances. Rapids’s government can’t even give them a straight answer on whether an NDA exists. They’ve said on their website no NDA has been signed, but wouldn’t give a yes or no answer at a planning meeting.

One of the objectives of the organizing in Rapids is to make the unanswered questions obvious, and give available information to the community.

Kaitlyn: Yesterday, we walked in a parade, just a small parade outside of town here. And we had flyers, half-page sheets that we were handing out. And I was saying, here’s a page of info because we’re willing to inform you. Because the city’s not. They’re not.

Emily: Also, people are allowed to take our information and say, ‘I don’t agree. I want it to be built.’ That’s fine. They’re allowed to look at all of our information and make that decision for themselves. I will say, nobody showed up to the planning commission meeting to say they wanted it. Everybody who spoke didn’t want it. Everyone said, ‘I don’t want it,’ or, ‘I have huge concerns about the way the city has been handling it, so that makes me not want it to come to town, because I feel like you’re hiding something about it.’ Those kinds of things. People asked, rightfully, about the moral character of Eric Adams. He has federal indictments for bribery and he wants to be financially involved in a project here. I think that’s a bad idea.

Disgraced former New York City mayor Eric Adams showed up at a Wisconsin Rapids town hall earlier this month to advocate for the data center. His reception was not warm. He reportedly said to the crowd, “Like it or not folks, data centers are going to be here,” to which a resident in attendance yelled, “Put it in your backyard!”

Kaitlyn: I know people who have come to the conclusion that they are against this project because of how the city has handled it. Especially once the whole Eric Adams thing happened, and when it comes out that these are Russian billionaires who are owning this — there are citizens looking at that and saying, ‘I’m really not anti-AI or anti-data-center, but this is really not the right project.’

Emily: I would really love to know how many gallons of city water they’re going to be using, so I know how much my bill is going to go up when it happens.

Kaitlyn: They want clean water for that closed loop system. How are they going to get clean water other than drawing from the city? That site is so heavily contaminated from being a paper mill that they’re not supposed to touch that land. You can’t dig at that site. And if you do put tanks in the ground, they’re next to the Wisconsin River. What happens when they get a leak? There’s so many unanswered questions.

Emily: They won’t tell us what chemicals will be used in the close loop system. That’s an industry secret. They are adamant that it will be closed loop, And because of that, there will be no contamination. In Wyoming, a Meta data center using a closed loop system put antibacterial resistant bacteria into the city water supply. Because before you start the closed loop system, you have to– it’s called a fill and flush. And when they flushed it, all of the bacteria built up in those pipes went straight to the city water supply.

Danielle: They won’t tell us how often they’re gonna flush that system. They really haven’t answered a single question.

One way the city has tried to compromise with residents is proposing a Community Benefit Agreement. Residents are highly skeptical that the city would be willing or able to enforce one — especially if PNK signs it, but leases the center to a different tenant. Even if the tenant was legally bound to the agreement, and the city properly enforced it, would punishment in the form of fines even matter?

Emily: In my mind, a bunch of billionaires who aren’t from here would rather pay a $10,000 fine every day for the rest of their life than actually adhere to the code or the ordinance, because they have the money to do that. So why would they follow it if they can just pay what is really a slap on the wrist fine for them for the rest of time?

Danielle: I live about a mile downwind from the dog food plant, and they have the same thing. They have scrubbers for the smell and everything. Some days I can’t even be outside because it smells like rotten garbage in the sunshine. It is the most awful, awful smell. And I have complained for years about it. They’ve fixed it and there’s fines. Well, they pay their thousands of dollars of fines, and it still stinks to high heaven for three days in a row.

So I’ve got that on this side of me, and now, less than a mile on this side I’m gonna have whatever’s coming out of [a data center]. How are you going to enforce it? And if the city gets them to pay a fine, they’re going to pay the city. They’re not paying me. You know, the city’s going to get that money. I don’t know, in a sense, getting paid for screwing up my life.

This is where you naturally start to wonder what the pros of this proposal even are; at least, what the city and PNK are trying to claim they are. Hypothetical tax benefits are the reason Wisconsin’s Republican-controlled state legislature and Democratic Governor both okayed a massive sales tax exemption three years ago to lure data centers to the state in the first place.

An actively functioning data center wouldn’t come close to replacing the hundreds of jobs the mill provided; residents know whatever big tech company moves in likely won’t hire locally much. They found out during one recent public forum that even temporary construction jobs to build the data center wouldn’t be hired in Wisconsin Rapids, either.

Danielle: At the meeting at the library, they said that they were going to build it off-site and bring it in. So it’s going to be a modular building anyway, kind of like a Lego situation.

Kaitlyn: They want to have a bunch of it pre-built and they’ll just slap it together on the land.

The loss of the mill was economically devastating for Rapids, but it also meant the end of over a century of pollution on its site. They shared with me that they’d lost loved ones to illnesses believed to be related to the mill’s pollution. Public and private officials can’t or won’t give residents answers about whether their safety would be at risk. They can’t tell them if the center would bring local jobs, or how many.

Emily: Let’s enter a hypothetical: The data center will not do any kind of pollution. It will do nothing bad. It will increase the tax break or the tax base or whatever the hell. But. because it uses a lot of power, your electricity or water or whatever bills are going to go up. There’s nothing you can do about it. Everybody’s bills are going to go up. No other harmful environmental impacts. Whatever, whatever. We’re in a hypothetical.

There are people. 40% of the people who live in Wood county are in the ALICE population. That extra $20, $30, $40 a month is the difference between them having fresh fruits and vegetables or not. It’s the difference between the lights being on or the kids having WiFi to do schoolwork. That $30, $40, $50 is a lot of money to people around here.

I also struggle with that. If the economy is already struggling and 40% of people around here are employed and still living paycheck to paycheck, that extra $50 of the energy bill every month is massive.

Danielle: And people will say, ‘Oh, Rapids runs everything out. Rapids doesn’t want anything but the mill.’ We go by South Wood County Neighbors for Responsible Development. Not because we don’t want development, but because we don’t want damaging, extremely detrimental, irresponsible development.

Kaitlyn: There’s nothing with a data center coming that doesn’t look bleak as hell.

The sentiment among Rapids residents is seemingly unanimous. I asked if anyone in city government has taken a public stance against the data center. They said no, not publicly.

Kaitlyn: Because they don’t want to get sued.

Danielle: For having a bias. If PNK isn’t able to build here because of anything, they can take whatever words that any Alderman or anybody has used.

Kaitlyn: ‘You had a bias against us, and that’s why you didn’t approve our conditional use permit.’

Subscribe now

Does PNK, or any billion-dollar company, get to bully their way into Wisconsin Rapids, public sentiment, safety, and environmental impact aside? And with environmental impact: where is the Department of Natural Resources?

Answer: Former Governor Scott Walker’s administration slashed their budget. Wisconsin, birthplace of 20th century environmentalism, has faced some of the nation’s most vicious cuts to its environmental agencies in the last fifteen years.

Kaitlyn: The DNR does not have a very strong public statement on data centers. In the beginning I called the DNR and I said, ‘Could you make a statement on this?’ The local DNR guy called me back and said, ‘I will email you our fact sheet.’ And that was basically it. The DNR frustrated me in the beginning a lot. And I haven’t gone down that path more, because I was like, I can not deal with you if you’re just going to waste my time.

Wisconsin Rapids is a red city in a red county. They’re on the same side of the data center issue as the Party for Socialism and Liberation. The ideological pluralism behind this issue hasn’t seemed possible in a long time in Wisconsin.

As a member of the Milwaukee PSL I’ve spoken with put it: “Data centers and Flock cameras. They’ve brought everyone together.”

Kaitlyn: This has crossed political lines extremely.

Danielle: It’s mended fences even in my family.

Emily: I have been really surprised about how it’s crossed party lines in Wood County. The demographic of people that showed up at the first planning commission meeting very much surprised me. And I think this data center conversation is a great way to start mending some of those relationships. We do agree on things. We might not agree on how to get there, but this is where we can start.

Kaitlyn: This is all from my opinion, where the city is coming from. This is all about tax revenue, and this is all about this potential tax revenue. Because they lost the mill. I don’t know how much debt we as a city carry now, but we do not have the tax base to do the things we need to do to maintain this city.

Which I understand. But you don’t bring in someone who says you may get a billion dollars in tax revenue. We’re just speaking in hypotheticals constantly.

The fight to bring data centers into Wisconsin makes my chest hurt. The fight to keep them out, the more I learn about it, gives me hope in more ways than one. It’s reinvigorating local democratic participation in Wisconsin, and overruling partisan lines.

The organizers feel that optimism too — even through being called terrorists and anonymous threats online. One had security cameras installed recently because of backlash. I asked if she feels safe. She shrugged, and said, “These are Russian billionaires we’re talking about here.”

Emily: We’re happy to volunteer our time for this, because we care about the community and don’t want it. But every time we learn something, it gets even further entangled with a different city and a different person and different company.

Kaitlyn: I personally have optimism in the discussions that I’ve had with the lawyers. There is plenty of fight left to be fought. This is absolutely not a done deal.

Emily: And nobody is safe from a recall at the moment. I think they all realize that based on the vibe in that crowd, none of them are safe from community action.

Three final questions, for now. First, has anything in recent memory brought Rapids residents together like this?

Danielle: I’m thinking over my history. Some of the things that, you know, made people up in arms. Back when we had monkeys at the zoo, there was a lady who activated a bunch of people. She thought it was very inhumane. We never had monkeys at the zoo again. So that was the last major controversy.

Objectively excellent and unexpectedly fun answer.

Second: What do you need from the state of Wisconsin?

Emily: I need [Gov.] Tony Evers to reverse his tax incentive for data centers to come to town. His signing that was detrimental to Wisconsin in my opinion. I’m really bummed that the gubernatorial candidates, neither of them are interested in a moratorium or a pause on data centers to learn more about them. That’s a really big bummer to me. I wish there was a candidate, either side of the aisle, who was at least talking about it.

Kaitlyn: I’ve basically given up on the higher government. The mayor gave an interview where he kind of blames the state for not doing anything.

Finally: Who do you feel is listening?

Emily: That’s a good question. I think there are residents listening to us. They might not take everything we say immediately. They might read it and then do some more research, whatever way that sways them. But I think there are people who are appreciative that we are doing something, and they are willing to read our posts or listen to what we say or say.

Kaitlyn: I think other cities are watching us. I know Stevens Point is watching us. I would think that Marshfield is watching us. I think that we are going to be the example for the cities in this area.

Emily: I mean, people in New York are watching us. Eric Adams came to town and now we’re on the news in New York.

Someone in San Francisco once dryly called my Wisconsin ID “exotic.” I thought about that as I drove home. They were joking, obviously, but there’s an uncomfortable truth at the heart of it. What are places like this to places like that?

Someone yelled at Eric Adams to build a data center in his backyard, if they’re so necessary. Those who stand to profit from data centers will continue to lobby for their imposing, destructive existence anywhere that they don’t personally have to experience.

Wisconsin Rapids isn’t the first or the last town to be reduced from a community and culture to land for big tech to put things they don’t want to see. And they aren’t the first or last to take that control back.

–

Other things I’ve paid attention to this week:

Jason Kelce in this eccentric musical number asking people to send jars of pee to data centers:

At one of the tech companies that I worked for lifetimes ago, for example—a blockchain startup likely committing fraud—all of our work was in service of the software developers. We were shown videos of sweat-stained Microsoft CEO Steve Ballmer chanting “developers, developers” to motivate us.

In that era, it seemed that not having people skills or emotional intelligence was actually perceived as a mark of genius.

All pendulums that swing one way are bound to swing back the other, however! This week, a viral excerpt from a Financial Times article caught my attention. In the excerpt, a “financier” speaks to the FT journalist about vetting new hires. Specifically, vetting recent young entrants to the job markets who have never known work without AI.

The employer told the reporter that while the “AI natives” he initially interviewed appeared “wildly impressive,” they were actually “alarmingly shallow.” “Critical thinking,” he continues, is the most important hireable skill—a skill that humanities graduates seem to have more of at the present than STEM ones.

This employer is not the first to speak of humanities as a more marketable skill in this new age. A few months ago, Netflix co-founder Reed Hastings admitted that in the age of AI, we will all seek emotional experiences from humans—-that no one wants to watch a bunch of robots play basketball.

Even Daniela Amodei, the co-founder of Anthropic, claimed in an interview a few months back that Anthropic looks for people who “are great communicators, who have excellent EQ and people skills, who are kind and compassionate and curious and want to help other people.”

For all the talk of “getting left behind if you don’t use AI,” an emphasis on the humanities right now actually makes complete sense. Because when you have a tool such as AI that can predict the predictable parts of a job, it’s unfortunately the preference of some current employers to employ the predictive tool rather than to employ you.

What humanities teaches you, though, is how to adapt to the less predictable—given its inherent study of the inherently unpredictable human being. It allows for acknowledgment of the spaces in-between that require nuance and context, context that AI frequently lacks. Because for all the rare books AI can shred and read, it is quite literally not able to read a room.

For instance, take a person who needs to think swiftly in a meeting based on the unpredictable actions of another human being. That person’s behavior could be the difference between landing a contract or not. In countless ways, there is simply no replacement for the intuition of a human, with a library of physical and emotional lived experiences.

This is not to say that people in STEM can’t solve complex problems—nor that AI can’t try to emulate human intuition or originality. It just means that an understanding of the unpredictable human psyche will be more coveted when “predictability” is abundant. That we’ll see people who produce work without a huge AI footprint as having conscious and problem-solving minds—minds able to create new pathways for creativity, boundary-crossing, and connection-making.

If you look at the needs a lot of employers have these days—especially tech companies—they are rooted in skills of persuasion and communication. This is in part because the general public is not responding to data centers and late-stage capitalism. It’s also because we can sometimes tell if something feels unoriginal or inauthentic. I’ve never met someone who felt good about consuming slop.

Speaking of unoriginal: it’s also becoming uncool in the culture to use AI for everything. Gen Alpha even coined the phrase “that’s so AI” in reference to someone who is unoriginal.

Will there be a “humanities” bootcamp, an analog to those coding bootcamps that sprung up to train people for entry-level software developer roles? Probably not. (In part because a “humanities bootcamp” is a bit oxymoronic.)

But at a very high level, there may be a merging of the left and right brains—like a course I recently read about at the University of Colorado Boulder that is co-taught by a mathematician and a Renaissance scholar.

Because I’m not sure that STEM people and humanities people should exist in competitive silos, especially in tech. Perhaps instead of hiring one or the other, these people should be partnered together.

Subscribe now

What else we’re paying attention to…

Investigative outlet 404 Media placed an Airtag in a rare book that they suspected was on the path to training AI. The book was traced to an Amazon facility where they shred the books and feed them into large language models. (On the door of the Amazon facility was a Tyrannosaurus Rex eating a book.)

AI companies, realizing they are not very popular across political affiliation and demographic, are trying a new line of messaging: to pitch themselves as the cure to cancer.

Google bought all of Spirit Airlines’ data for $10 million to train its AI.

Walmart workers are correcting AI errors: “An online fulfillment worker told Business Insider she hasn’t been able to pick general merchandise orders as quickly as the AI app says she should, in part because of the AI-generated route through the store.”

This was not a bold take on my part, nor has it come true, at least in the way that I foresaw. A few congressional bills that would significantly rein in the worst excesses of prediction markets—especially sports betting, which makes up a sizable percentage of “trades” on these platforms—have not gotten much traction yet. There also hasn’t been a singular, glass-breaking scandal this year. Americans are gambling quite a bit, and the World Cup was unbelievably lucrative for both Kalshi and Polymarket. A recent poll even showed that a significant portion of Gen Z is deliberately moving investment funds from stocks to sports bets. Presumably, this has been made much easier by Robinhood, which has an app that allows users to easily toggle between investing in Apple and “trading” on “sports events.”

Despite growing revenues and a lack of congressional oversight, I’d argue the last handful of months have actually been pretty terrible for the main prediction markets, which are failing to fend off a painful death by a thousand cuts. They, and their Trump allies at the Commodity Futures Trading Commission (CFTC), seem legitimately spooked by an onslaught of lawsuits. Google seems to be spooked too, though its quiet retreat hasn’t gotten nearly as much attention as its announcement last year that it was integrating prediction market data into its AI-enhanced “Finance” tab.

Cities and states are suing prediction markets, mostly because of the sports bets that Polymarket and Kalshi actively advertise. As I’m writing this, news just broke that the city of Baltimore is suing both Kalshi and Polymarket, alleging that they’re operating as unlicensed gambling companies. (This is on top of a separate lawsuit filed by the state of Maryland, which is ongoing.) Minnesota came dangerously close to becoming the first state to ban prediction markets outright, until a federal judge issued a preliminary injunction a few weeks ago. On July 31, the state of New York sued Kalshi; State Attorney General Letitia James said in a statement, “No matter what they call themselves, prediction markets like Kalshi are gambling platforms, plain and simple.” New York City Council has since announced a separate investigation into “deceptive and predatory marketing practices by prediction market platforms,” according to a press release.

The cherry on top: 44 state attorneys general—forty-four!—complained to the CFTC last month that it is exceeding its mandate by attempting to regulate sports betting on prediction markets. Above all else—the addictive nature of gambling, safeguards for minors, the societal rot of being able to wager on basically everything—blue and red states alike are displeased that they aren’t collecting taxes on prediction market bets placed within their boundaries. They are right to be displeased! For all of the sportsbooks’ many faults, they are, at least, taxed on their revenues by states that have chosen to permit DraftKings, FanDuel, etc.

Kalshi and Polymarket have mostly hidden behind the CFTC and a defense that they are federally licensed exchanges not beholden to individual states. Like most Trump-friendly entities, the CFTC isn’t very convincing. On Aug. 11, the agency exercised its “emergency authority” to keep Kalshi operating in New York. This was done in the spirit of “market stability,” as if the entire Kalshi ecosystem would collapse because, during the dog days of summer, guys in Queens could no longer place “event contracts” on Da Freakin New York Mets. I am only barely kidding: in a statement, CFTC Chairman Michael Selig referred to Kalshi as one of multiple “interstate financial markets,” dramatically adding, “New York intends to make event contract derivatives waste away under its iron curtain of state gaming laws before the courts get the chance to issue final rulings.”

With the CFTC’s protection, prediction markets are going to squeeze as much juice as they can out of sports bets, as well as novel forms of degeneracy like gambling on flight delays. (FlightAware actually sued Kalshi this week for using its data and trademarks, then withdrew the suit, perhaps because Kalshi removed references to FlightAware and added disclaimers about its data.) I’m sure there will be excitable headlines this fall about all the money that prediction markets are racking in from NFL games and midterm elections.

But look beyond the betting volume. The lawsuits speak for themselves; they’re existential. The CFTC, which is a lackey for the prediction markets so long as Trump is in office, is interspersing dramatic statements about a gambling “iron curtain” with (mealy-mouthed, but still!) concessions meant to eliminate some of the stupidest “event contracts.” A Wall Street Journal exposé about false, deceptive advertising from Polymarket was damaging enough to compel a supposed company audit, plus an anonymous admission from the CFTC that it’s conducting an investigation. In response to New York City Council’s probe into the marketing practices of prediction markets, Polymarket issued a surprisingly conciliatory statement: “We look forward to engaging with The New York City Council on this matter,” a spokesperson told CNBC.

You can also intuit Kalshi and Polymarket’s precarious position through Google’s lawyer-speak and the fact that its prediction market integration has been a dud. We’ve come a long way since November 2025, when the illegal search monopoly slash journalism-killer touted a new feature where users could open up the Google Finance tab, “ask questions about future market events” (referring to Kalshi and Polymarket), and “harness the wisdom of the crowds.”

In April, Futurism reported on how Polymarket bets were showing up in the Google News tab next to actual news stories. Google issued a statement chalking up the Polymarket promo as an error. Then, sometime between June and July, Google added a skittish disclaimer about its use of prediction markets data. “This data is sourced directly from third-party market providers and is not generated or calculated by Google,” the disclaimer reads. It also includes a section on “regulatory and financial risks,” including that “the regulatory framework surrounding prediction markets is complex and continues to evolve,” so “you are responsible for understanding and adhering to the regulations applicable in your specific location.”

That’s not all! As of August 1, the Google Chrome Web Store outright banned extensions that “facilitate or enable real-money transactions on predictive outcomes.” The Chrome Web Store didn’t differentiate between sportsbooks and prediction markets: “We don’t allow content or services that facilitate or promote real money gambling or prediction markets, including but not limited to online casinos, sports betting, lotteries, or games of skill that offer prizes of cash or other value.”

Last night, I messed around with Google Finance’s main search tab, the real hub for this prediction market partnership. I was hoping to pull up some sports and midterm election “event contracts.” Wouldn’t you know it? No results. I was only able to surface forecast modules for “event contracts” outside of sports, politics, and culture. I tried to chat with Gemini about this, which repeatedly returned error messages and gave conflicting information about why the Finance tab wasn’t prominently displaying sports or election events via Kalshi and Polymarket.

Best as I can tell, Google has been filtering out controversial sports and election markets from its Finance tab. It probably has been doing so for a while, if not the entirety of their partnership. On one hand: good! Makes sense! On the other: not a ringing endorsement of the primary use case for Polymarket and Kalshi.

As cities, states, and media outlets continue to sue and scrutinize prediction markets, I suspect Google and other initially proud allies will further distance themselves, so as to avoid additional reputational harm. Including, one can only hope, Kalshi’s top talent: Timothée Chalamet. What the hell, man?

Here’s what else we’re reading this week:

• Wired and Mission Local reported on two different AI-run websites that purport to be journalistic ventures.

  • One of the outlets, RuntimeWire, has actually scored a few scoops by ingesting transcripts and court documents, then publishing news stories, faster than humans can. Most of the articles I scanned on RuntimeWire are bland and poorly structured, and the site’s creator doesn’t exactly have a foolproof plan for editing, fact-checking, and legal reads. (In short: he uses AI agents and his own judgment.) But I can at least see the usefulness of having AI tools constantly scanning for newsworthy transcripts and filings.

  • The other AI-run site highlighted by Wired and Mission Local is The Dissent. I do not like it. Every story has a fake byline, so as to assign an anthropomorphic quality to the AI agents who are (badly) aggregating actual journalism. The creator of The Dissent did not consult with any journalists about this pet project, according to Mission Local.

• Xavier Becerra has been quiet since clinching the Democratic Party nomination for California governor. He made a rare public appearance at a Politico-sponsored event earlier this week, and demonstrated why others in his party do not speak very highly of his message discipline and political vision.

  • Becerra floated something or other about offering shares in AI companies to Californians, then walked it back, while adding that such an idea is, ultimately, not “totally far fetched.” So which is it? I suppose we’ll find out eventually.

  • Becerra also lamented how “we hardly have any” AI regulations in California, which is true, though California still far outpaces other states.

  • And Becerra stepped on the toes of Rob Bonta, who succeeded him as state attorney general, by suggesting California’s lawsuit to block the merger between Paramount Skydance and Warner Bros. should end with a settlement.

• Rep. Ro Khanna, whose district encompasses Silicon Valley, seems to be inching closer and closer to an official presidential run. Khanna recently swatted down a Big Tech-backed challenger in his primary, and told NPR that achieving “America’s next New Deal” will require standing up to tech oligarchs.

  • “These folks would have been the conquerors in a different era,” he said. “They kind of think of themselves as above you and I. And when I’m actually most inclined to run, I think, ‘I know these people.’ I understand the economic future, and I will have the guts to stand up to them.”

• Sam Altman told a group of tech interns the following: “I think we are close to a world where you can have, like, a descendant of ChatGPT watch your computer screen all of the time, watch every meeting you’re in, record every call, everything like that. Have perfect context of your whole life, everything you see. You choose what information you want it to have, but it can like—you can connect it to your texts or emails or Docs or Slack or whatever. And then you kind of have this thing that’s not making decisions for you, but if you’re the CEO of a startup, there’s always more stuff for you to do and context you can’t all keep track of. You can’t read every piece of customer feedback every day. You can just have this thing working alongside you.”

Alphabet, the parent company for Google, has been particularly aggressive in pushing that narrative. Alphabet President Ruth Porat said last fall that “[w]e should be able to cure cancer in our lifetime” thanks to AI. Demis Hassabis, the co-founder of DeepMind, the tech giant’s most accomplished AI lab, has gone even further, suggesting in a 2024 60 Minutes interview that “we can cure all disease with the help of AI.”

The emptiness of that rhetoric, always suspected, is now quite clear.

Two weeks ago, Google shut down the team responsible for AlphaFold, DeepMind’s AI-powered protein structure prediction tool. AlphaFold represented perhaps the most persuasive data point in favor of the narrative that AI could lead to scientific breakthroughs, including in cancer research. The model’s ability to accurately predict protein structures helped Hassabis become a Nobel laureate. A week after shuttering AlphaFold, Google removed Hassabis from his role directing DeepMind, using the time-worn tactic of giving Hassabis a face-saving “promotion” to a powerless role.

The Financial Times shed light on Google’s reasons for these moves:

While Hassabis is widely admired for his scientific leadership, several people familiar with Google’s thinking said senior executives had become frustrated by what they saw as his lesser focus on the commercial demands of the company’s AI business.

His decision to make AlphaFold . . . freely available became a source of tension inside Google, according to a person familiar with the matter. The project generated little commercial return despite the scale of its investment.

In other words, AlphaFold was too much of a scientific endeavor for Google’s liking. So Google shut the project down and kicked its key backer upstairs to a role where he couldn’t hold back the tech giant’s profits.

If someone had consciously set out to demonstrate the disingenuousness of Big Tech’s grandiose AI promises, they could hardly have scripted it better.

*************************************

AlphaFold’s protein-prediction capabilities arguably represent the most significant scientific achievement that can be attributed directly to advances in artificial intelligence.

Proteins consist of chains of molecules called amino acids. Many amino acid chains “fold” in characteristic ways to form the structure of specific proteins. Misfolded proteins directly cause certain diseases and are believed to play a role in many others, including Alzheimer’s, Parkinson’s, and certain cancers. Consequently, knowing the physical structure that results from protein folding is important for many areas of drug discovery and development. But scientists have long struggled to predict a protein’s three-dimensional structure even if they know the protein’s amino acid sequence.

Share

DeepMind designed AlphaFold to help resolve this prediction problem. The first version of AlphaFold won a prominent protein-folding prediction competition in 2018. Its successor, AlphaFold 2, blew away other entries at the same competition two years later, helping Hassabis become one of three laureates for the 2024 Nobel Prize in Chemistry. The last iteration of the model, AlphaFold 3, was able to predict more complicated structures and interactions involving proteins, DNA, RNA, and other molecules.

One would think that Google would have bent over backwards to boost AlphaFold. In addition to providing a rare example of a concrete AI-driven scientific achievement, AlphaFold offered an at least vaguely plausible path to the AI industry following through on its rhetoric about curing cancer through AI.

The company decided to instead kill the project because (1) AlphaFold’s breakthrough was not generalizable and, relatedly, (2) Google was struggling to profit from it.

*************************************

AlphaFold is an amazing tool, but it is still just a tool. While it represents a major technical breakthrough for the specific tasks it was designed to perform, it is of little-to-no use in unrelated (or, indeed, even in closely related) fields of science. The additional capabilities of successive generations of AlphaFold came not from AlphaFold improving itself, but from its (human) developers imbuing it with new capabilities by designing more sophisticated models with improved architectures and expanded training data sets. So while AlphaFold’s capabilities are impressive, they are also quite narrow.

On that note—AlphaFold did not “solve” the protein folding problem, as some news sources (including some science-focused outlets) suggested. There are a significant number of proteins whose structure AlphaFold cannot reliably predict. Even for protein structures it predicts with high accuracy, AlphaFold cannot explain why or how the proteins fold the way they do. (AlphaFold’s inability to generalize outside of the tasks it was trained to perform very much applies to large language models, as Gary Marcus frequently points out.)

Which leads to the second point. Alphabet is spending $200 billion in 2026 alone on new AI infrastructure. Because Hassabis insisted on making AlphaFold’s code and database publicly available, there was little prospect that the project would help Google recoup its enormous AI investments. Had Hassabis agreed to make the invention proprietary, Google would enjoy a commercial monopoly on its use. Because he didn’t, the technology’s commercial potential was limited. That, it seems, sealed the fates of AlphaFold and Hassabis.

In viewing cancer through the lens of profit maximization, Google is simply following in the footsteps of the rest of Corporate America. Myriad Genetics, a corporation based in Salt Lake City, infamously filed patents on a number of human genes that it discovered through novel testing techniques, including two genes linked to breast cancer, so that it could retain a monopoly on the market for selling tests for the gene. The Supreme Court thankfully ended that particular practice by ruling in 2013 that human genes could not be patented. Cancer drugs can, however, still be proprietary, a fact that drug companies have used to set the price for many patented cancer treatments at extortionate levels.

Against that backdrop, Google’s actions are not surprising. But they certainly are galling. Google, like the rest of the AI industry, had gone to great lengths to convince policymakers and the public that we all should get behind Big Tech’s vision of an AI future. It then sidelined Hassabis because (God forbid) he dared to take some modest steps that might have allowed others to have some small share in that future.

At least the charade can now end. Big Tech does not, in fact, care about curing cancer — unless it happens to boost profits. Let’s all remember that the next time a tech executive attempts to justify the AI industry’s actions by touting all of the scientific wonders AI will supposedly help humanity achieve.

Update (Aug. 14, 2026): The introduction was updated to clarify that Google shut down the AlphaFold project team, rather than the AlphaFold tool itself. The rest of the article, which correctly characterizes Google’s actions, has not been changed.

Matt Scherer is a fellow at Open Markets Institute, where his research and advocacy focus on developing policy responses to the eventual bursting of the AI bubble. His Hard Reset pieces focus on highlighting the risks posed by the AI bubble and pushing back against the hype that is inflating it. The opinions expressed here are solely his own.

It’s wild to think about such a salary range for mid-level corporate roles—especially when many people can’t even get an interview with an AI “recruiter” because of how many qualified people are applying for one job, and how skewed the labor market currently is in favor of employers these days.

This is not to say that tech workers should be underpaid, or not paid handsomely. (Anthropic is valued at $965 billion.) But when I read that Dario Amodei, the CEO of Anthropic, was concerned that people were joining his company for the salary and not the mission, it did make me wonder: how is a company that is compiling billions of data points on human nature so confused about human nature?

The tech labor market is not exactly ideal right now. Layoffs abound, because everyone from software engineers to marketing managers are either being “replaced by AI” or AI is being used as the excuse to wipe away those jobs. Even the people conducting the fires are said to be informed by AI’s analyses and suggestions.

In other words, to be gainfully and securely employed by a tech company today is, in many ways, to work for an AI company. Especially for employees who haven’t yet experienced a blockbuster IPO—there’s no denying that Anthropic’s upcoming IPO probably is the golden carrot of generational wealth taunting them to join over anything else.

To Amodei’s point, though, I’ve seen several people across various social feeds announce that they are joining Anthropic for “mission-driven” or for prestige reasons. One woman who sold her online harassment mitigation startup announced recently that she was joining Anthropic. Presumably she doesn’t have to work for money, and is interested in being at the epicenter of the “least evil” AI employer.

It also may be that recruits like her want to be close to one of the most consequential technological revolutions in human history. They may feel excited to influence and build from the inside—to have an “impact” on products that hundreds of millions of people may be using, carrying an ethos that may be adjacent to effective altruism. Perhaps they also see it as a mark of their own superintelligence to be recruited to build superintelligence. (Speaking of human nature, ego is a thing, too!)

Anthropic is also a public benefit corporation—meaning that its governance structure is built to prioritize long-term societal gains over short-term investor gains. So it follows that Amodei wants to hang his hat on attracting the people who believe they have a moral compass. I have to believe that he is drinking his own Kool-Aid, that he and his intelligent band of less-than-evil employees will be better than the alternative.

However: Anthropic, to be competitive in the AI ecosystem, is still beholden to interests that require them to pour an unbelievable amount of money into AI infrastructure. Anthropic, in order to compete in the same game as OpenAI and others, must still participate in the strategies of the game in which everyone else is operating.

For example, although it was a brilliant PR move, we’d be naive to think that Anthropic—which was providing its technology to the U.S. government’s Iran war efforts—didn’t want to steal some users from ChatGPT when it vocally declined to work with the military.

So yes, sure: recruits to Anthropic could rationalize that this is the less evil option, that they are doing this for humanity.

But as a former tech worker myself, I’ve seen people move from one hyped tech product to the next: ridesharing to scooters, crypto to artificial intelligence. And while AI is far more consequential in its macro and micro implications than some of these circle-jerk products like crypto, the AI train may be perceived by some tech workers as the last money-making opportunity to benefit the insiders before even more job losses occur. For someone like Amodei who is compiling one of the largest databases of neural networks in human history, to miss this insight seems like a massive blind spot.

What else we’re paying attention to this week…

Mayor Mamdani is supporting a union-backed bill that would require Amazon to hire delivery workers rather than relying on contractors.

Is it the end of Google search?

A PR firm composed of AI avatars is sending out pitches to journalists. The firm’s clients said they weren’t aware of the practice.

Mark Zuckerberg is at it again with a 6,500 word manifesto making the case for AI superintelligence—and predicting that we’ll all have “AI agents” predicting all of our wants and needs in the near-future. I enjoyed this commentary piece about it entitled Mark Zuckerberg doesn’t understand how to live.

Speaking of Zuck, Meta has pledged to invest $1 billion in communities where it’s developing and operating data centers; meanwhile, they are investing $600 billion in data centers by 2028.

There are escalating levels of AI dependency. Overreliance is, frankly, the best-case scenario. The situation gets more dire when users incorporate AI into most of their real-life interactions, or come to believe that their dozens of coding agents unlock supercomputer-levels of productivity. The Wall Street Journal just reported on how tech entrepreneurs and venture capitalists are increasingly recording their meetings and even their dates, so that conversations can be distilled and analyzed by Claude afterwards. This bucket of folks, unsurprisingly cloistered in Silicon Valley, thinks everyone else is weird for finding their behavior to be concerning. And then there’s a batch of users who suffer from life-threatening delusions as the result of their extensive, near-constant dialogue with chatbots. In some cases, these users have allegedly been driven to self-harm and violence.

It seems Hank Green, a world-famous YouTuber with millions of followers, has an unhealthy relationship with an AI chatbot. Luckily for him, he was called out for his behavior last week, while he was still in the “overreliance” phase. His full response to this controversy wasn’t necessarily laudatory—I found it borderline-cloying and definitely annoying—but the fact that he ultimately fessed up and admitted he has a problem is, in my opinion, a first-of-its-kind development and a blueprint that shouldn’t be outright ridiculed.

For the uninitiated: Green is basically Bill Nye for Gen Z. For decades, he has enthusiastically educated students (and curious adults) about science, the environment, and all sorts of other pet interests. His brother is John Green, world-famous author of novels including The Fault in Our Stars. They are both YouTube pioneers and have squeaky-clean reputations, especially compared to every other internet-famous figure with a parasocial following.

Long story much shorter (Vulture has an explainer if you want the nitty-gritty details), Green was recently criticized by his fans for using an AI-generated diagram in an educational video called “You Have Never Seen a Normal Star.” The diagram apparently had incorrect information in it, among other issues. Green acknowledged inserting the AI-generated diagram and agreed it was in bad taste.

Green’s fans remained on high-alert about his AI usage, which he’s been teasing out with increasing frequency on podcasts and social media posts. Then last week, when he was reading off a script during one of his videos, he tossed in the phrase, “And I appreciate the pushback.”

It was an odd interjection, and the exact genre of groveling that chatbots use when you tell them they got something wrong. Fans asked him what was going on, and at first, Green obfuscated. In a now-deleted X post, Green wrote that he produced the video in question “under a ton of pressure” and “did use ChatGPT for research on this script,” adding, “watching it I definitely get an AI feel, so I think it’s fair to say I was relying too much on generated notes.” But he also claimed to have ad-libbed the “I appreciate the pushback line.” If he was lying about that, yikes. If he was telling the truth, then it means he’s emulating the cadence of a chatbot, which is probably worse.

The latter scenario suggests Green is spending a ton of time talking to ChatGPT. He’s hardly alone among power users. Google’s AI “users are writing queries three times as long as the keyword-heavy questions they asked of traditional searches,” the New York Times reported, citing Google’s own research. “People are spending one to nine more minutes in AI Mode than on traditional Google searches,” according to the Times. That’s just Google Gemini, arguably the least “charismatic,” most annoying of the mainstream chatbot options. ChatGPT and Claude offer an even more personalized experience, and an alarming number of people enjoy that intimacy: 16% of Americans, including 25% of users under 30, utilize chatbots for “companionship,” according to a 2025 poll from the Associated Press.

After deleting his X post, Green re-emerged on Reddit with another explanation and something approaching an apology. “I’m mortified that I have let so many people down and I think I understand how that happened,” he began. He promised to “change things about how I make stuff” and admitted that he’s been “diluting” himself by “relying too heavily on AI as a research aid.” He’s clearly in the bargaining phase, because he wrote multiple paragraphs about how he’s more bullish about AI than other people in his community. Eventually, though, he wrote the following:

“But mostly I need to come to terms with the fact that the level of dopamine I’ve been getting from interacting with LLMs...with doing more and more and more and more...is not healthy for me or good for the world. It is careless, and has disconnected me from where people are on this.”

He invoked his wife and brother, both of whom “did not pull any punches” when they talked to him about his AI dependency: “Neither of them think [I’m being] healthy,” he wrote. “I need to readjust.” He ended his note by announcing he’s going to pull back on some of his work, and that he’s “very sorry for harming the thing this is all based on.”

Transparently, I find Green’s AI reflections to be fairly unsympathetic and grating. His line about “dopamine” hits from LLMs is silly and dramatic. (Hank, you’re using AI to research and possibly script episodes of your show; you are not Robyn.) Green’s assertion that reasonable people can disagree over whether AI is going to destroy the fabric of society also drives me nuts. It reads like fence-sitting and cope—a defense of his behavior and beliefs only after he got caught in the act.

I struggle to sympathize with anyone, not just Green, who’s exposed for depending so heavily on AI. I harshly judge journalists and writers who lean on AI to draft their work—it’s a duplicitous, oft-disqualifying practice. Tech executives who are haughty about their extensive AI usage are even worse. I totally get why Green’s fans are unsettled and angry by his belated disclosure.

All that said, I see a silver lining here. Green might not be querying ChatGPT all day for a “dopamine hit,” or whatever, but he certainly appears to have experienced something approaching an addictive relationship with his chatbot. He is hardly alone. As much as I want to roll my eyes at his (and others) predicament, I don’t think that’s always an appropriate or constructive response.

The chatbot users who end up falling victim to delusions and worse were, at one point, lost but not gone. Their chatbot was an unhealthy companion, not yet a total life-ruiner. They could’ve benefitted from an intervention of sorts. Green got one on the internet and, it sounds like, from his family.

Was there too much handwringing when Green finally came clean? Yes. His message is still a net-positive. Green is, at least as far as I can tell, the first legitimately famous person to come out and say they’ve become too reliant on AI. It’s a meaningful concession, and undoubtedly an affecting message for others in a similar position.

After months of chatter about tokenmaxxing, amassing armies of AI agents while running on no sleep, and how every successive model released by Anthropic is more powerful than God, it’s nice to witness a cringey-but-earnest reality check and reset. I have no idea if Green will stick to his pledge, but I’m going to try and take him at his word, if for no other reason than he might just help other people snap out of it.

Here’s what else we’re reading this week:

• Regular people who learn about and/or encounter Flock Safety tend not to like it. The mass surveillance company is aware that its invasive technology is difficult to defend, and so it is reportedly providing talking points and guidance to police departments. Those talking points include the following: “Don’t avoid the concept of mass surveillance because you’re not going to convince opponents that it’s not.”

• Nikita Bier, the head of product at X, announced he’s stepping down. He took a victory lap about his 13 months on the job, during which the X platform somehow became more unusable.

• In Michigan’s 7th Congressional District, William Lawrence scored a hard-fought victory over two other Democratic Party primary challengers. Lawrence, a progressive, has called for a year-long federal moratorium on new data centers. “Tonight the people of Michigan’s 7th District sent a clear message: they are tired of Silicon Valley billionaires and DC insiders determining the future of our towns and our country,” he wrote in a statement after his win on Tuesday.

• Francesca Hong, the frontrunner in Wisconsin’s Democratic Party gubernatorial race, has also made a data center moratorium a central part of her campaign. The Wisconsin primary is next week. Putting my pundit hat on: voters will care more about her position on data centers than her grouchy holiday takes.

• Speaking of data centers, I (like many others) enjoyed Jasmine Sun’s writeup of her trip through Wisconsin and Michigan, where she talked to a bunch of data center critics and proponents and unpacked why this has rapidly become such a potent political issue.

• Anduril Industries co-founder Palmer Luckey is now the “godfather of Republican fundraising on the West Coast,” a GOP operative told The Guardian.

The opening statement is succinct and digestible, and I encourage you to take a moment to read it, but its six points of belief in summary are:

1. AI is a socio-political technology.

2. AI should be governed as a normal technology.

3. AI’s trajectory is not predetermined.

4. To be critical of AI is to be reasonable.

5. To be critical of AI is to question power and information asymmetries.

6. The most important AI harms are currently hidden.

The rapid and unregulated development of AI, the dominant narrative of its inevitability, and its effects on labor, environment, war, and government surveillance — civilization, as one co-signer summarized — have converged in an urgent way this year. From a university famous for its status in Silicon Valley, an open call to action on AI is significant. Its creators and co-signers are highly aware of this. They say Stanford’s longtime influence in technological development comes with responsibility to criticize the trajectory and consequences of those technologies.

I got to speak with four co-signers and one co-founder of the Manyfesto, all from different research backgrounds and Universities. Here’s what they said.

Conversations edited for length and clarity.

Chijioke Mgbahurike, a Law & AI Research Fellow with Stanford’s Liftlab, is one of seven co-founders of Stanford Critical AI, which officially launched in March. His research focuses on sociotechnical implications of AI bias on marginalized communities. He told me the project has been in development since last October, when he and other co-founders attended a speaker panel about AI and warfare, and came away from it frustrated.

Share

“I think the main impetus was the feeling that Stanford was not very critical towards AI in ways that we found satisfactory, and that rather than waiting for someone to change that culture, that we can just change it ourselves,” he told me. “I know some popular figures think about [criticism] like AI populism, or like the backlash towards data centers, as this group of people who are under-informed… I’ve seen a lot of these narratives come through both through Stanford and the Bay more generally… When I was at that particular panel, I could appreciate the analysis, but I realized in that moment there was nothing new that I didn’t already know. And I thought that I was kind of looking for action as opposed to some kind of analysis.”

As for why the group decided on a manifesto, Mgbahurike said the political nature of this kind of document rises to the political nature of AI as a technology, and of Stanford’s place in its ecosystem.

“We must recognize that, with the sort of spread of tech manifestos, what we’re seeing is this combination of both politics and technology,” he said. “We need to, as a community, take a look at what our institutions created, and what we ourselves want to create in that world.”

Co-signer Dr. Lucy Suchman is an anthropologist who has had a critical engagement with artificial intelligence for over 40 years. Her research has focused on, among others, the politics of technology design and militarization of AI. She told me there’s a direct throughline from concerns in the 1980s about the automation of control systems for nuclear weapons to the presence and use of AI in militaries today.

“I’ve always been interested in the social, political, economic relations that give rise to these technologies and that inform the way that they’re developed. And so now we’re seeing those political economies in particular reaching their sort of apex,” she said. “They’re sort of playing out to their logical extensions as venture capitalist speculative investments.

“Consistently, those who are most prominent in relations, particularly to government inquiries into AI and its regulation and its problems, are the same actors. We have the military commercial complex and the revolving door where people are constantly moving between government and private sector positions. The idea is, well, we’re talking about these technologies. We have to get the people with the most expertise. It turns out that they’re the same people who are running these companies. And so they end up on the boards that are advising the government. They end up effectively writing the legislation, having enormous influence. And this absolutely marginalizes, and in many cases, kind of shuts down other voices.”

Dr. Wenshan Jia is a communications scholar who researches the impact of technology on international intercultural communication. He told me studying AI through that lens is incredibly unique to traditional media.

“This is something which tries to emulate, even replace human agency,” he said. “It’s very unique in that ChatGPT or large language models were developed without the input of a peer review process… The development of the large language models itself was developed with the money from the Corporations, and it’s really close-ended, no peer review process. So therefore, when they developed it, there was nothing in the process of making sure that they’re ethical. And so as a result of that, they create lots of problems. And most importantly, technologically speaking, we are at this narrow intelligence period, but the next stage is artificial general intelligence, and the third stage is artificial super intelligence. And most importantly, humanity will be enslaved.”

He’s concerned about human agency in the age of AI, and the survival of critical thinking in young people who come through school using generative language models to produce work for them — but as the Manyfesto itself concurs, he is not against the use of AI overall, even within academia. As a communications researcher, he says that developing a media literacy framework specific to AI is crucial.

“AI literacy is quite different than social media literacy, than internet literacy, or media literacy in general,” he said. “I think [the general public] need[s] to be informed of the potential consequences of the massive use of AI in all aspects of life. They need to be providing guardrails, they need to be given options when to use it, how to use it. If AI is present, they’d be aware of it and how to reduce it, how to develop a kind of a healthy relationship between AI and them. And that’s why I think AI literacy is very significant. It should be a kind of education which cuts across all society sectors.”

Dr. Andrés Burbano studies the history of media technologies, and is involved in a historical initiative to track the evolution of the critique of artificial intelligence. He told me the impact on distribution of political information in the age of AI and social media, and its effect on democracies everywhere, should be one of the priorities of our times.

“These transformations are affecting all of us, every person with a mobile phone in their hands. But even beyond that, people that don’t have access to communication systems are affected by these transformations,” he said. “We need to have more information about the history of how the field took shape as a technology, AI, as a set of technologies. But also to identify that critical thinking and a critique of this kind of technology has been present for a long while. That is not something that the present brings.”

Dominic Zappia, a research associate with Stanford’s Technology Ethics & Policy Rising Scholars Program with a background in economics, philosophy, and political science, told me he sees how AI is regulated and who regulates it as a reflection of the wider sociopolitical moment.

“I think that it’s furthering this general sort of sense that people have that they’re not in control of what’s going on. The feeling of being left behind embodied in the populist movements of Trump,” he said. “I would argue that people feel very little positive benefits from AI in their daily life. And if they are positive benefits, they’re diffuse and the harms are very concentrated. The harms of AI diffusion are the data centers. They’re people losing jobs. So when you see this thing happening and you feel like it’s bad, and there’s nothing you can do about it, that continues to erode trust… And there’s also general resentment against Big Tech. It’s bad, is the takeaway. And I think that Silicon Valley doesn’t feel that as much as it should.”

The Manyfesto is a call for open criticism of AI as a method of intervention. Its interdisciplinary backers highlight, through the lens of their different areas of research, that AI as a technology, as well as the pace of its development and exponential prominence in everyday life, is everyone’s problem. Questioning AI’s inevitability and its presence in any given space is important, and not to be dismissed as “technophobic” or anti-progress. Everyone I spoke to emphasized that the general public, as powerless as we can feel over AI’s takeover, retains this ability and responsibility to question. Academia, and the Manyfesto, is simply a framework for criticism.

Share Hard Reset

“My hope has been that when people find out about our group, a lot of people are surprised that this could come out of Stanford,” Mgbahurike told me. “One of the goals of the work is that, ‘you guys are not alone, you guys are not powerless’. There are people who are fighting for the basic dignities that we believe you are entitled to. And that just because you cannot make it into kind of these, quote unquote—I use this as a derogatory term—hallowed spaces of Stanford or any of the higher level institutions does not mean that your ideas, your beliefs and your wants no longer matter.”

Stanford Critical AI says the project will aim to build community, advance AI research, and create spaces for advocacy on campus and beyond, to reiterate AI criticism as a contemporary civic and academic responsibility. The Manyfesto and its co-signers want you to know, no matter what is framed as inevitable: the future of AI is still a matter of public choice, and the public is entitled to questioning it.

The background of the clip was as such: onstage at a conference, in a sort of fireside chat, Altman is asked about AI safety mechanisms. His response is that if we overreact and implement too many AI safety mechanisms that centralize AI under one control, humans will live “in service” of AI, and that we will have no privacy or agency. Essentially, he says, we’ll live in a surveillance state (of course I, Sam Altman, won’t surveil you—but the government or one of my competitors might.)

Several people criticized my reposting of this clip, claiming that I was “fear-mongering” and that I was sharing Altman’s response “out of context.” These respondents assured me—with a bizarre sort of moral authority for people who were defending AI oligarchs—that Altman was warning us of what could happen, not that he wanted it to happen.

He was merely responding to a question about the most terrifying AI scenario, they said—not promoting it.

Putting aside the fact that Altman has blown through all kinds of safety guardrails in his relentless pursuit of building OpenAI—for example, allowing advertisers to mine ChatGPT conversations, as well as building a relationship with the Department of War—I found it fascinating that people were obsessed with honoring the specific interview context of Altman’s response.

Because the context of Sam Altman—reported in several investigations, and through many testimonies—is that he is a power-hungry person, without a lot of regard for ethics, willing to steamroll anyone in his way. (As one board member put it to New Yorker writer Ronan Farrow: “I don’t think Sam is the guy who should have his finger on the button.”)

So with that context, it makes it difficult for me to believe that Altman was just innocently responding to a question about AI’s worst case scenario. It makes it impossible to think he is just a concerned individual wanting to ring the alarm about the surveillance state.

Because even if his response was stated in response to a posed question, Altman most certainly—in building what he built to begin with—must have already been considering people being spied upon or losing their critical thinking skills as a potential outcome. It didn’t terrify him enough, however, not to pursue it.

And say you could plausibly believe or argue that Altman was trying to warn us of becoming slaves to the AI. That he was, in good faith, trying to propose the worst case in order to prevent it from happening. That he wasn’t, like all of his AI competitor CEOs, trying to position himself as the one genius positioned to save us all.

Even if all of that were the case, how much stock should we put into any person who is posing as a savior to the very problem that they had a hand in creating?

And the idea of people losing their personal agency to AI—as he lays out in the worst case—is not actually a Black Mirror hypothetical. It’s already happening.

Case in point: the other day, Altman posted, in earnest, a “positive” use case for ChatGPT. The use case he proposed was: kids and parents can use ChatGPT to create podcasts about their days—feeding calendar information about soccer games and music lessons into the model so that it can create a machine-narrated version of storytime.

It’s hard to imagine something that strips a person of their agency more than reducing children’s experiences into a calendar-prompted machine-narrated podcast. Beyond limiting open and free-flowing communication between parents and their young children, this dynamic also zaps the creativity that so naturally springs from kids.

So yes, context is important. And that context may very well be that Sam Altman doesn’t find the worst case scenario so dystopian after all.

What else we’re paying attention to…

New York sued Kalshi, alleging it is an “illegal gambling operation.” The attorney general said platforms like Kalshi, Coinbase, and Gemini can “encourage problem gambling, including by people under age 21, and endanger ​people’s financial, emotional and physical health.”

Even babies are not free from surveillance tech. From a deep-dive into the baby monitoring industry: The comedian John Mulaney recently riffed on his toddler son’s sleep setup, saying his bedroom looked like a “C.I.A. black site,” with a blackout tent over his crib. “I don’t know why we have 24-hour CCTV footage of my son’s crib, he’s not going anywhere — he sleeps in a bag,” he said. “We then have three sound machines going at full blast. And because we read it once on Instagram, we keep his nursery at 68 degrees. Any warmer and your infant will die automatically.”

With the rise of vibe-coding, startups are seeing million-dollar valuations with just one employee.

Scammers in Cambodia were using ChatGPT “to create fake online personalities, translate conversations, draft messages to potential victims, and produce promotional material. Some users also generated forged passports, legal notices, stock purchase confirmations, and fake cryptocurrency trading interfaces.”

Uber’s Strategy for Fighting Sexual Assault Suits: ‘What Were You Wearing?’ Gross.

A former operations specialist in the tech industry describes the challenges of reentering tech in a first-person essay, I spent a decade being loyal to tech in Ireland. Then AI took my job: “To make matters worse, the interviewers rarely help you show your true potential. While some were kind, many others treated me with condescension. They looked down on me from a position of absolute power. They treated the interview like a cold interrogation rather than a mutual partnership.”

Share