Hello! My name is Kelly Stonelake. I worked at Meta from 2009 until 2024, including as Director of Product Marketing. After experiencing sexual harassment and witnessing child safety failings, plus Meta’s retaliation toward women speaking up about it, I filed a lawsuit and became a federal whistleblower and advocate for reform. I write this Overturned newsletter on tech accountability and broken systems of power.
This morning, Meta announced a landmark settlement with 47 states, the District of Columbia, and U.S. Territories where they will pay up to $18 billion over the next 10 years to abate damages allegedly caused by their products, and make agreed platform changes to protect young people.
I’m pleased to see a meaningful financial settlement, but I’m much more interested in the product changes Meta has agreed to make: limits on teen use, overnight restrictions, fewer notifications, stronger age assurance, new parental controls, and the ability to choose a chronological feed. Meta has consistently disputed claims that its product design contributes to harms to children. Now, after years of litigation and with a federal trial already underway, it has agreed to legally enforceable restrictions on many of those design choices.
Meta is accepting a two-hour default daily limit for teens, nighttime restrictions, limits on notifications during school and sleeping hours, controls over autoplay and algorithmic feeds, stronger age-assurance requirements, and independent auditing of its compliance.
These are concrete interventions, and it should not have taken this much to get here.
It took a bipartisan coalition of attorneys general, years of litigation, repeated attempts by Meta to defeat the case, and ultimately a trial in federal court before Meta agreed to these changes. When Meta presents this settlement as evidence of its leadership on child safety, we shouldn’t accept this narrative. These protections were not simply volunteered by the company as the next evolution of its safety program. They were negotiated under extraordinary legal pressure.
I was surprised to see Meta Chief Legal Officer C.J. Mahoney saying in the company’s announcement that “this framework will only work if all our peers join us” and that “we need an industry-wide solution.”
I agree with him; kids deserve meaningful protections regardless of which platform they happen to be using, and I hope this becomes an industry-wide floor. But it is remarkable to hear that argument coming from Meta.
This is a company that has categorically denied the allegations against it, disputed claims about the harms associated with its products, fought efforts to hold it legally responsible, and repeatedly argued that its existing safety efforts were substantial. Now, with a major trial underway and billions of dollars at stake, Meta is not only accepting significant restrictions on its products, it is calling on the rest of the industry to follow them.
For years, technology companies have asked policymakers and the public to trust them to identify problems, balance safety against other considerations, and make changes voluntarily. Yet some of the most consequential protections in this agreement arrived only after sustained outside pressure. To me, that is powerful evidence of the limits of relying on the companies themselves to decide what level of protection children deserve.
And that is why the independent auditing provisions may ultimately matter as much as any individual product change. Meta has announced child-safety tools and protections many times before. This agreement creates an independent mechanism for testing whether Meta is actually complying. Accountability cannot consist of Meta announcing a feature, publishing a blog post about how well it works, and then asking parents and regulators to take the company’s word for it.
The implementation, auditing and enforcement of these changes will therefore be critical. We need to know whether Meta accurately identifies young users, whether the protections actually reach them, whether the company complies with the spirit as well as the letter of the agreement, and whether these interventions meaningfully change children’s experiences and safety on its platforms.
If Meta believes these measures are necessary enough to become an industry standard, then the obvious question is why children had to wait until attorneys general took the company to federal court to get them.
Meta spent years insisting that the people raising these concerns were wrong, that its products were safe, and that it could be trusted to police itself. Today, it agreed to change those products and declared that the rest of the industry should do the same.
The problem was never that Meta didn’t understand the harm caused by their products, or that Meta could not make safer choices. It was that, until now, it had too little incentive to do so.
How Oakland connects directly to what I saw inside Meta:
Related Interviews:
There’s so much fear underneath the statement that this will only work if peers join them. It indicates a belief that these protections designed into the product are a business risk rather than a differentiator. Maybe these design changes will make Instagram a preferred social media platform for parents and teens. Rather than seeing this as an untapped opportunity in the market, one there is demand for, they are fearful of losing what they’ve already built. That stance is so contrary to the early days of working there and what made the company successful in the first place.
I wish adults can choose to have a chronological feed too