Instagram's head, Adam Mosseri, appeared in federal court in Oakland, California, on Tuesday, becoming the first company executive to testify in the lawsuit brought against Meta by 29 states. He spent roughly an hour on the witness stand and is expected to continue on Wednesday. California, Colorado, Kentucky, and New Jersey allege that the company designed Facebook and Instagram to be addictive for young users while misleading consumers about safety. The 29 states jointly accuse the company of violating the federal Children's Online Privacy Protection Act by collecting and using personal information from children under 13 without obtaining parental consent as required. The states have indicated they may seek nearly $200 billion in civil penalties.

Meta denies intentionally making children addicted, stating that internal research has not established a clear link between teen usage and diminished well-being. The jury will provide an advisory verdict, while presiding Judge Yvonne Gonzalez Rogers will determine liability, penalties, and whether to mandate changes to Facebook and Instagram's product rules. The trial is expected to extend well into September. The case is People of the State of California v. Meta Platforms, U.S. District Court for the Northern District of California, Oakland Division, case number 23-cv-05448. Mosseri, 43, has led Instagram since 2018, and Mark Zuckerberg is expected to testify later. Mosseri acknowledged that some safety tools have not changed how long young people spend on the platform. Internal documents described teen safety tools as a 'proof point' that were 'not effective' at reducing related behaviors. Colorado prosecutor Jason Slothouber asked if he was surprised by this; Mosseri said he was not.

The 'Take a Break' feature launched on Instagram in late 2021, prompting users with a reminder to close the app after a set period of usage. A blog post from that time stated that over 90% of teens who enabled the feature during early testing kept it on. However, internal materials presented in court showed that only 1.8% of teens ever actually used the feature. These two figures measure different things: one reflects retention among those who turned it on, the other reflects adoption across all teens. The blog highlighted the former but omitted the latter. Slothouber asked whether parents had any way at the time to know that only about 1% to 2% of teens would use the feature. Mosseri answered, 'Correct.' When asked if the company had ever publicly disclosed that percentage, he again answered, 'Correct.' He noted that he had publicly stated the 'opt-in rate was low' but said he was 'not sure if the specific number was ever reported.' 'We don't publish every statistic,' he added, while also saying the prompt was 'helpful, but not nearly as helpful as we would like.' 'Most teens didn't want it. We decided to push forward anyway.' In 2024, when the company introduced teen accounts, it made the feature enabled by default for teens, nearly three years after its launch. He denied the states' claim that the company deliberately delayed making it the default.

Mosseri listed other notification-limiting tools, such as Quiet Mode, as part of a broader array of safety initiatives. Product design director Francesco Fugu testified earlier, stating that he held monthly safety feature meetings with Mosseri. Mosseri framed safety tools as strengthening year over year, with more parental controls and additional teen restrictions. The states contrasted this with internal admissions that such measures do not reduce behavior, painting a picture of tools marketed for safety while acknowledged internally as ineffective. Liability findings have not yet been reached. After lawyers removed figures from a slide, Mosseri said it was news to him. A 2023 internal presentation noted that content related to suicide, self-harm, and eating disorders reached teen audiences at about 2.5 times the rate of adults, and that content labeled 'not recommended for teens' was viewed by teens at about 1.5 times the intensity of adults. Company lawyers subsequently asked employees who created the presentation to remove these figures to 'limit' Mosseri's 'exposure to such information.'

Mosseri testified that he had never participated in conversations about restricting his own access to information. Such arrangements were 'news' to him. 'I'm not a legal expert, and I don't know why materials were removed before reaching me, but I have not encouraged my team to hide anything,' he said. 'I want to understand how things work. I can't think of a time when I encouraged anyone to give me less information.' When asked whether lawyers reviewed child safety research for litigation risk, he said he would 'not be surprised' if it were for complex legal and policy reasons, but would be surprised if the intent were to hedge against litigation or mislead the public. An email from 2023, where an employee forwarded a lawyer's warning about 'litigation risk' regarding a teen safety presentation, was also shown in court. Following Frances Haugen's leak of internal documents in 2021, Mosseri said the company became more cautious about sensitive research. He disagreed with characterizing the public messaging as deceptive, stating that safety information shared with the public should be high-quality and expert-backed. He agreed that the company should not lie to users. When Slothouber asked whether his frequent public appearances were aimed at 'personifying' Instagram, Mosseri said explaining the company's actions to the public was 'important and valuable.' These are witness statements, not a verdict.