The first week of one of the most consequential tech trials in US history wrapped up in Oakland, California, with prosecutors boiling Meta's alleged playbook down to four words: hook, hold, harvest, hide.
The owner of Facebook and Instagram "hooks" in users, "holds" them on its platforms for as long as possible, "harvests" their data and then "hides" the truth from the public, Megan O'Neill, a lawyer for the state of California, told the jury. "Meta's business model worked especially well for kids," she said, according to The Guardian's report from the courthouse.
The blockbuster trial opened on Tuesday in Oakland, just north of Meta's Silicon Valley headquarters. California has joined 28 other US states in suing the roughly $1.36 trillion company, alleging it designed addictive products that harm children and violated federal child privacy laws and state consumer protection statutes by collecting data on children under the age of 13 without parental permission. The case is being closely watched across the AI policy world as a stress test for whether engagement-driven product design can be legislated out of existence — a debate covered in depth in AI industry analysis.
A Star Witness With Internal Data
The most dramatic testimony of the trial's first week came from Arturo Béjar, a safety engineer who worked at Meta in two separate stints between 2009 and 2021 and has since become one of the company's most prominent internal critics, testifying before a US Senate committee and serving as an expert witness in other social media cases.
Béjar told the jury his motivation for pursuing child-safety solutions was personal: his own teenage daughter received unwanted sexual advances, photos of male genitals and misogynistic insults on Instagram, and found the platform's reporting processes either ineffective or impossible to use. "Meta is taking a 'don't ask, don't tell' strategy" when it comes to child safety, he testified.
His testimony carried unusual weight because of how close he once was to the top of the company. Béjar said his job regularly included briefing Mark Zuckerberg, and that he had spoken with the CEO more than 100 times. During his testimony, attorneys for the states showed the jury an email Béjar sent Zuckerberg in 2021 outlining a survey of teenagers' experiences on Instagram. The results were stark: 51% of teen users surveyed reported bad or harmful experiences within the previous seven days, and harmful content was taken down only 0.02% of the time.
Béjar said he sent the data to the top because, in his words, "when Mark makes something a priority, mountains move." Asked whether Zuckerberg ever responded, he replied: "No. I didn't hear back from him."
Meta fought hard to keep Béjar off the stand, filing a series of motions to strike his exhibits and block his testimony. All were rejected.
Meta's Defense: An 'Outlandish Payout'
Meta has denied all allegations. Company spokesperson Liza Crenshaw said the states had decided "to chase an outlandish payout" rather than stick "to the facts or the law."
During opening statements, Meta attorney Paul Schmidt conceded there is "no dispute" that people can struggle with social media, but argued the company has built tools to address those harms. He noted that Meta does not allow children under 13 to register accounts and said it has disabled more than one million accounts belonging to underage users.
The jury also heard from other former employees and a psychologist during the first week, and eight jurors will continue to weigh evidence in a trial expected to last six to eight weeks.
What's at Stake: Up to $200 Billion
The threat to Meta is existential. If the company is found liable, damages could reach as high as $200 billion — roughly equivalent to Meta's entire 2025 annual revenue, as The Guardian noted. The states are also asking the court to force Meta to redesign its products to make them safer for children, a remedy that could permanently reshape how its platforms operate.
One unusual wrinkle: the jury's role is advisory. The panel will issue recommendations, but Judge Yvonne Gonzalez Rogers, who is presiding over the case, will make the final decision on the verdict and damages. Proceedings are being led by attorneys for California, Colorado, Kentucky and New Jersey, and the jury is expected to eventually hear testimony from Meta CEO Mark Zuckerberg and Instagram CEO Adam Mosseri.
The Oakland trial is the largest of thousands of similar US lawsuits Meta faces from families, school districts and other attorneys general. The company has already lost the first two of those cases to reach trial: in March, it was ordered to pay nearly $1 billion to the state of New Mexico over child sexual exploitation on its platforms, and in a separate case it was found liable for deliberately designing addictive products that hooked a young woman, with damages of more than $4 million.
Why This Trial Matters Beyond Meta
For the technology industry, the case is a landmark test of whether the law can reach the design decisions behind engagement-optimized platforms. Prosecutors are arguing that maximizing time-on-site and data collection from minors is not just a business choice but a legal liability — and that internal knowledge of harm, of the kind Béjar documented in his 2021 email, makes companies accountable for it.
A verdict forcing design changes would set a precedent reaching well beyond one social media company, potentially influencing how every platform that algorithmically ranks and recommends content approaches young users. With lawmakers in Washington and Brussels still wrestling with how to regulate AI systems that curate what billions of people see, the Oakland courtroom has become the most important stage in tech policy this fall.
Stay Ahead of AI
For the latest developments in AI policy, platform regulation, and industry analysis, bookmark AI Buzz Wire.
Read more AI news →
