Multi-state youth safety trial against Meta opens with claims of hook, harvest, hide

August 18, 2026, 22:08 GMT | Insight Hook, hold, harvest, hide. These four words will anchor the case a group of state attorneys general will make to a federal judge and advisory jury over the next six weeks: that Meta Platforms designed its platforms to encourage compulsive use by young people while misleading the public about the risks.

In her opening statement Tuesday, California Deputy Attorney General Megan O’Neill said these four words make up Meta’s business model: “Hook the users. Hold them for as long as they can. Harvest their data, hide the truth from the public when making public statements,” she said. “And Meta's business model, it worked especially well for kids.”

She said the jury will hear how Meta “hooked kids on its platforms” and designed Facebook and Instagram to keep them coming back. “You're gonna hear how Meta desperately tried to hold on to those kids. How it wanted them so badly. How it wanted them young,” she said. “The young ones are the best ones. That's the title of an internal Meta study we're gonna show you.”

The 29 states, led at trial by California, Colorado, Kentucky and New Jersey, are also claiming that Meta violated the Children’s Online Privacy Protection Act by knowingly collecting and using personal information from children under 13 without parental notice and consent.

O’Neill said Meta needed kids’ data to make money just as much as it needed to reassure parents and the public that kids were safe on its platforms. “Meta repeatedly went out and said the platforms were safe for kids. You'll hear that Meta realized it had a PR problem and it needed to get ahead of that problem by pushing out messages about safety, especially to parents,” she said. What followed was a steady drumbeat of statements and talking points: that Instagram and Facebook were safe for kids, that Meta didn’t design its platforms to hook children, that Meta prioritized safety over profits and that Meta kept users under 13 off Facebook and Instagram.

O’Neill showed the eight-member jury a slate of disillusioned former Meta employees who will testify to these allegations, including former director of engineering Arturo Bejar (see here),  former head of youth policy Vaishnavi Jayakumar (see here), former user experience researchers Jason Sattizahn and Natalie Troxel (see here and here), and former vice president of product design, Margaret Stewart.

Nearly all of these employees have testified already — either in a previous case brought by the New Mexico attorney general (see here), or in the first personal injury trial against Meta and YouTube over social media addiction (see here).

One notable exception is Stewart, though she testified by video Tuesday at a trial currently unfolding in Tennessee (see here).

The Oakland jury will also hear Stewart’s video deposition.

Meta CEO Mark Zuckerberg and Instagram CEO Adam Mosseri are also expected to testify in person, along with Antigone Davis, Meta’s head of global safety (see here). Both Zuckerberg and Mosseri testified live in Los Angeles (see here and here).

The four lead states also plan to call expert witnesses like Dr. Mitch Prinstein, the chief science officer of the American Psychological Association, and psychology researcher Dr. Jean Twenge to bolster their claims about the effects of prolonged social media use on young people’s mental health and developing adolescent brains.

O’Neill also pointed to Meta’s own internal research. Showing the jury the cover page of a 2020 Meta study titled “Teen Fundamentals” that featured an image of a young brain, she said: “For the past decade and more Meta has used these features to draw kids onto the apps and keep them coming back. To hook them.”

“You're gonna hear about how Meta knew a lot about kids' brains. How they are constantly seeking rewards, how they're so sensitive to social feedback, how they're still developing their ability to control impulses, the way adults do. Just think about the teenager in your life, right? You're gonna hear that Meta researched these aspects, these vulnerabilities of the teen brain,” she said.

She also assailed Meta’s repeated assertions that Facebook and Instagram are not designed for kids, displaying a snippet of a study titled: “Long term retention: the young ones are the best ones and other learnings,” that said “Tweens (approximate age 10-12) are special. People who join Facebook as tweens have the highest long term retention out of all age groups.”

According to the states’ expert Carl Saba, there were 46 million instances across the four states over a 10-year period in which a child spent at least 60 hours a month on a single app. He also found that from 2012-2022, the four states had 9.7 million teen Instagram users and 6.9 million teen Facebook users. There were also 4.6 million children under 13 on Instagram and 3.9 million children under 13 on Facebook.

Meta strongly disputes that it turned a blind eye to these young users. Meta attorney Paul Schmidt said that the evidence will show that between 2020 and 2024, Meta disabled 1,481,631 accounts it “checkpointed” and deemed to belong to users under the age of 13

“This is acting in a meaningful way to remove under 13 users,” he said.

Schmidt noted the extreme difficulty of verifying users’ ages, especially when account holders lie about how old they are when they create an account.

“Some of the most frequently reported under 13 accounts belong to fully grown adult celebrities,” he said. “People don't like the account and it's reported as under 13.”

Meta now requires users to enter their ages when they set up accounts on either platform, he said, “but that's not the end of Meta’s efforts. It continues after people sign up. It uses computer detection and review. It allows people to report accounts of people under 13, and it reviews those accounts. It trains its reviewers to look for other signals. If they see an account and they think it might be under 13, it trains them to act on it. And if they find one that they think is under 13, they do what’s called checkpointing. They stop collecting data and they investigate.”

He also challenged the states’ assertion that Meta launched a coordinated campaign to mislead parents and the public about safety. Schmidt said out of all the statements Meta employees have made around the world, which number in the tens of thousands since 2012, “they have pulled out a tiny subset of those statements that they have identified to us and have said they are false. We will ask you to consider two questions. First, was it actually false or was it based on facts and was it honest opinion? And two, did it matter? That requires you to consider the context of the statement.”

In many instances where the states cited Meta’s internal emails, group chats and other documents, Schmidt said the comments were taken out of context or instead reflected employees working to improve the company’s products and respond to findings from its own safety research. “I don't think you'll see any evidence of any other social media company doing that,” he said.

Meta will also emphasize the unsettled science around social media and mental health, pointing in particular to a National Academies consensus report that found social media use can be associated with both positive and negative outcomes for adolescents.

Schmidt said even the states’ own public health officials did not attribute youth mental health problems entirely to social media. He displayed for the jury a July 2025 quote from Michelle Barnes, director of the Colorado Department of Human Services, who said, “We had regular check-ins with the attorney general’s office because they are our lawyers and occasionally the attorney general himself. I have absolutely no recollection of them ever mentioning Facebook, Meta or Instagram.”

Perhaps most significantly, Schmidt said, was what the states left out. “In that list of words you were given there was one that you might have expected to hear that you didn't: Harm,” Schmidt said. “They're not going to bring people from their states to say ‘I suffered harm from these features they're talking about. I was deceived by these statements.' They're not going to bring in parents to talk about that.” He added that the states themselves won’t talk about how they suffered harm.

But the relief the states are seeking on behalf of their residents is significant. The potential for $1.4 trillion in penalties has been widely reported by the media as one of the key features of the case. But state attorneys general have retreated from that figure, which was offered as one of several alternative penalty calculations in disgorgement charts unsealed last month (see here).

Meta highlighted the $1.4 trillion figure, noting that it "alone is larger than the gross domestic product of most countries, even without considering the other penalty and disgorgement claims the AGs seek to add to the $1.4 trillion."

The states have walked back that number. “We have not made a request for $1.4 trillion,” said O’Neill at a hearing last week. She added, “We’ve obviously put forward different theories for how to calculate penalties, and we’ll have to see how the evidence comes in, but we wanted to give the court some comfort about where we are.”

The trial is expected to run through mid-September and is being overseen by US District Judge Yvonne Gonzalez Rogers (see here), who has presided over some of the biggest tech trials in history, including the recent blockbuster legal brawl between Elon Musk and OpenAI CEO Sam Altman (see here) over Musk’s claims that the company abandoned its non-profit status and breached its charitable trust to donors.

Gonzalez Rogers will be the final arbiter of whether Meta violated state consumer protection laws and the federal children' s privacy law, and if so, she will determine the penalties.

Please email editors@mlex.com to contact the editorial staff regarding this story, or to submit the names of lawyers and advisers.