You know how even if a boxing match is lopsided, both opponents will prance around afterwards acting like they won? The guy who got smacked around for 12 rounds tries to show the judges (and the crowd) that everything went according to plan, even while he’s got a visibly dislocated shoulder and a broken nose.
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.
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.”