Frances Haugen Warns About AI 'Friends' and Legal Battles
Facebook whistleblower Frances Haugen told El Pais on June 7, 2026 that "we are worse off today than when I leaked the Facebook documents" in 2021, warning that AI companion apps used by minors will be the next major legal battlefield after two landmark U.S. verdicts against Meta. A Santa Fe jury ordered Meta to pay $375 million for misleading consumers and enabling child exploitation on March 24, and a Los Angeles jury found Meta and Google liable for $6 million in a bellwether case tied to roughly 2,000 pending suits over addictive platform design. Haugen, who left Facebook with 21,000 internal documents and now runs the nonprofit Beyond the Screen, argues the same design-liability arguments used against social feeds will extend to AI chatbots marketed as companions to children.
The legal playbook behind this spring's landmark verdicts against Meta and Google, targeting addictive product design rather than harmful content, is the same playbook Haugen says will soon reach AI companion apps. For teams building conversational agents accessible to minors, the design-liability precedent set in Los Angeles and New Mexico this March is now a more concrete risk model than content-moderation law alone.
What happened
Frances Haugen, the engineer who disclosed the Facebook Files in 2021, told El Pais in an interview published June 7, 2026, "We are worse off today than when I leaked the Facebook documents." She said the next major legal battle will target the AI companion apps children use, extending the same accountability arguments now landing against social platforms. El Pais reports Haugen left Facebook with roughly 21,000 internal documents, testified before the U.S. Senate, and later founded the nonprofit Beyond the Screen.
Legal context
On March 24, 2026, a jury in Santa Fe, New Mexico found Meta liable under the state's Unfair Practices Act for misleading consumers and enabling child sexual exploitation on its platforms, ordering the maximum civil penalty of $375 million, according to the New Mexico Department of Justice. New Mexico Attorney General Raul Torrez called it the first time a state has prevailed at trial against a major tech company over child harm; a bench trial beginning May 4 will decide additional public-nuisance penalties and whether Meta must overhaul age verification. A day later, on March 25, a Los Angeles jury found Meta and Google (YouTube) liable for negligent platform design in a case brought by a young woman who began compulsive use at age six, awarding $3 million in compensatory and $3 million in punitive damages (Meta responsible for 70%), per NPR. That case is a bellwether tied to roughly 2,000 similar pending lawsuits; both companies have said they will appeal.
Industry context
These verdicts build on a wave of litigation Haugen's 2021 disclosures helped trigger, including an October 2023 lawsuit filed by attorneys general from 41 states and Washington, D.C., alleging Meta designed Instagram and Facebook features to addict children. The plaintiffs' legal strategy in the Los Angeles case, framing platforms as defectively designed products rather than policing content, sidesteps Section 230 protections that have long shielded platforms from liability over user-generated content.
For practitioners
If courts and regulators extend defective-design liability from social feeds to AI companion products, the operative risk shifts from content moderation to provable safety-by-design: engagement mechanics, personalization logic, and age-appropriate guardrails become discoverable evidence. Teams building agents marketed to or used by minors should expect scrutiny of retention-optimizing design choices and should maintain documentation of safety tradeoffs made during development.
What to watch
The May 4 New Mexico bench trial phase, which could add injunctive requirements like mandatory age verification; Meta and Google's appeals in the Los Angeles case; and whether plaintiffs' firms already litigating the roughly 2,000 pending social media cases begin filing similar defective-design claims against AI companion or chatbot products aimed at young users.
Timeline
The Wall Street Journal begins publishing the Facebook Files, based on roughly 21,000 internal documents Haugen supplied.
Attorneys general from 41 states and Washington, D.C. sue Meta, alleging Instagram and Facebook are designed to addict children.
A Santa Fe jury orders Meta to pay $375 million for misleading consumers and enabling child exploitation in New Mexico.
A Los Angeles jury finds Meta and Google liable for $6 million in a bellwether addictive-design case.
Haugen tells El Pais the next legal battle will target AI companion apps used by minors.
Key Points
- 1Haugen said the next major legal battle after Meta's social-media verdicts will target AI companion apps children use.
- 2New Mexico and Los Angeles juries ordered Meta and Google to pay $375 million and $6 million over child-harm design claims.
- 3Design-defect liability theory used against social feeds could extend to AI chatbots, raising documentation and safety-by-design stakes for practitioners.
Scoring Rationale
Verified and substantially enriched: the vague 'recent courtroom developments' framing now cites the exact, confirmed $375M New Mexico and $6M Los Angeles verdicts with primary/independent sourcing (NM DOJ, NPR), and the Haugen AI-companion warning is a credible extension of an active, well-evidenced litigation trend rather than speculation. Kept in the notable range rather than major since the AI-companion litigation itself is still prospective, not yet filed.
Sources
Primary source and supporting public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
