Landmark Meta Settlement Targets Harmful Product Design, But Leaves Much on the Table

Washington, D.C. – A landmark settlement announced this week between Meta and 51 attorneys general would require meaningful changes to the design and operation of Instagram and Facebook for young users, together with up to $17.1 billion in financial relief, marking an important shift toward addressing harms through product design changes rather than relying on financial penalties alone. This agreement represents a key milestone in the long, hard-fought legal battle to make meaningful improvements to youth experiences online, which has united bipartisan AGs from nearly every state and territory in the US. Thousands of additional social media lawsuits remain pending.

The agreement requires protections for users under 18, including limits on daily use and nighttime access, hidden like and reaction counts, restrictions on notifications during school hours, limitations on account visibility, deletion of accounts for users under 13, and age assurance. It also establishes independent auditing and limited public reporting of Meta’s compliance.

When this agreement goes into effect, teens across the US will experience mandatory daily use limits on Instagram and Facebook for the first time, and Meta will be obligated to detect teens’ attempts to circumvent those limits. Yet the algorithmic feed – the core feature that will shape what teens experience during their limited usage time – is largely untouched by the settlement. Instead of being required to design feeds that best serve young users, all Meta has to do is send them a reminder four times per year that they can choose a (potentially spammy or repetitive) chronological feed. This optional chronological feed has been deployed in the European Union for years as a result of the Digital Services Act, and has not resulted in measurable evidence of improvements to youth satisfaction or wellbeing.

“If product designs are contributing to harm, effective remedies need to change those designs – not simply give young people another setting to manage,” said Peter Chapman, Deputy Director at the Knight-Georgetown Institute. “This settlement makes important progress by building protections like time limits and nighttime restrictions into the default experience. But leaving attention-maximizing feeds on by default puts the burden back on young people and their parents to opt into safer designs, which we know is not effective.”

That distinction matters. KGI’s Measuring Risk I: What EU Risk Assessments and US Litigation Reveal About Meta and TikTok report underscores how optional safety tools often reach only a small fraction of the young people they are intended to protect. Internal Meta data made public through litigation showed adoption rates below two percent for some safety tools and less than one percent of US youth users were enrolled in parental supervision. Safer options are most protective when young people do not have to seek them out.

The settlement appears to make some progress on governance and oversight, although many open questions remain. An independent auditor will assess Meta’s compliance and establish or approve the methodologies and metrics used to evaluate implementation, providing scrutiny beyond Meta’s own reporting. Getting those metrics right will be critical; effective oversight  should measure whether the required changes work in practice – not simply whether Meta has implemented them. Whether the auditor will have the access, mandate, and independence it needs to create real accountability is to be determined. The public will be able to see only non-confidential summaries of the auditor’s work, leaving the states to ensure their oversight has real teeth.

Meaningful oversight is central to Designing Technology Remedies, a framework developed by KGI, Tech Justice Law and the USC Marshall Neely Center for crafting effective and enforceable remedies for technology-related harms. The framework identifies three mutually reinforcing elements of effective technology remedies: harm prevention, which changes how companies design and deploy products; harm mitigation, which helps users avoid and respond to harmful experiences; and governance, which provides the oversight, measurement, transparency, and accountability needed to ensure remedies work.

The attorneys general settlement follows a recent New Mexico ruling that ordered changes to Facebook and Instagram together with nearly $1 billion in financial relief. The New Mexico remedy addressed some similar design changes while extending protections to AI chatbot features that are out of scope in this multi-state settlement, and focusing on child sexual exploitation and abuse. Together with enforcement actions being handed down in the EU and elsewhere, the glimmers of cross-jurisdictional accountability are beginning to take shape.

Yet this week’s settlement reminds us of the limitations of creating that accountability case-by-case. Meta’s mandatory usage limitations will last only five years, or up to ten years if Snap, TikTok, and Google agree to even stricter limits. This stands in contrast to the potential for more permanent impact that legislative approaches to the core drivers of harm could have. Similarly, the settlement’s age assurance provisions are highly specific to Meta and fail to account for some significant tradeoffs inherent in mandatory age assurance, including the balance between accuracy and feature availability for adult users.

Ultimately, the success of this settlement, and the resolution of the many pending cases in the pipeline, should be measured by outcomes – whether the new obligations meaningfully reduce the harms they are intended to address and make young people’s experiences online measurably safer.

Media contact: Julie Anne Miranda-Brobeck, Communications Director at the Knight-Georgetown Institute, jm3885@georgetown.edu