A New Mexico judge has found Meta a public nuisance. The company will hardly notice the bill

A NEW MEXICO judge on August 6th ordered Meta Platforms to pay $567 million into a fund to treat young people harmed by its social-media platforms. Combined with a $375 million in damages imposed by a jury in March, Meta's total liability in the case now stands at $942 million. The market response was barely perceptible: shares dipped less than half a percent to $589.44. That reaction tells the story. For a company that earned $60 billion in net income last year and commands a market capitalisation of roughly $1.5 trillion, the bill is a rounding error. The real question is whether the legal theory behind the ruling will prove harder to ignore.

Judge Bryan Biedscheid, a former public defender, went further than most courts have dared. He designated Meta a "public nuisance"—the first time a social-media company has received such a label. He compared Meta to a factory, in which advertising and content are the "product" and the psychological harm and sexual exploitation of children are the "pollution". The harm, he wrote, does not stay contained. It migrates to the internet and the real world, creating a "common, societal burden" on children, families, schools, hospitals and law enforcement.

The case was brought by Raúl Torrez, New Mexico's attorney-general, in 2023. It proceeded in two phases. The March jury verdict found that Meta had willfully violated the state's unfair practices act by misleading users about the safety of its platforms. The August ruling, reached after a three-week non-jury trial, addressed the public- nuisance claim. Judge Biedscheid concluded that Meta's platforms were a "significant contributing cause" of the mental-health crisis among the state's teenagers.

The judge's decree requires operational changes over the next five years. Teens in New Mexico will face a monthly usage limit of 90 hours (about three hours a day). Push notifications are banned between 10pm and 7am, and during school hours on weekdays. Like counts must be hidden by default for users under 18. Adults cannot message minors, and a one-strike policy will apply to adult users who engage in child sexual exploitation. Meta must prevent children from having romantic or sexualised interactions with its AI chatbots, and must develop an under-13 prediction model within two years.

Yet the court stopped well short of the changes that the state's lawyers wanted. Judge Biedscheid declined to alter algorithms, infinite scroll or autoplay, ruling that doing so could infringe on free-speech rights, harm Meta's competitive position or run afoul of Section 230 of the Communications Decency Act, which shields platforms from liability for third-party content. He also ruled that federal privacy law prevents Meta from being ordered to collect personal data for age verification in a way that would single it out from its rivals.

The result is a characteristically cautious intervention. The court has imposed safety guardrails and a bill, but it has left the core architecture of the business model—algorithmic engagement, the attention-extraction machine—intact. That is where the tension lies. If the harm comes from the way platforms capture and hold attention, then a ruling that preserves the attention mechanism while tacking on curfews and banners is akin to posting warning labels on a toxic product.

To be sure, the judge faced real constraints. Section 230 exists for a reason, and courts have struggled to distinguish between a platform's design choices and its editorial decisions. COPPA, the Children's Online Privacy Protection Act, was intended to protect young users from data harvesting, not to create a shield for predators. And the equity concern was genuine: ordering age verification only for Meta, while its rivals face no such requirement, would impose a competitive penalty unrelated to safety.

The deeper problem is that none of these constraints address the incentive structure. Meta's platforms are designed to maximise the time users spend on screen, because attention is the raw material of advertising revenue. Last year Meta earned just over $200 billion in revenue, the vast majority from ads. In the first quarter of 2026, that figure was $56 billion, up 33% year on year. Usage limits for a few million teens in New Mexico do nothing to change the calculus for the hundreds of millions of older users who drive the profits.

Meta will appeal, as it promised. The company argues that the ruling misrepresents the facts and that many of the ordered changes are "technologically impractical or completely impossible". Its broader litigation exposure is already significant. More than 40 states and over 1,300 school districts have filed similar lawsuits. A trial in Tennessee is under way, and a federal trial involving 29 states begins later this month in Oakland. The European Commission issued a preliminary finding in April that Meta has failed to prevent minors under 13 from using Instagram and Facebook.

Yet the domino analogy offered by Laura Edelson, an assistant professor at Northeastern University, may understate the structural risk. The New Mexico ruling matters not for its price tag but for its template. It established that public-nuisance claims can reach beyond the Section 230 shield when the allegation concerns the platform's own design rather than the content of third-party posts. It showed that a court can impose operational restrictions on a single state's users. And it created a factual record—Judge Biedscheid's findings on causation—that other litigants can point to by analogy.

The incentive misalignment runs both ways. States and school districts have a political motive to hold social media accountable, as youth mental-health outcomes worsen and parents demand action. But the legal theories they are developing are blunt instruments. Public nuisance was designed for littering streets and blocking harbours, not for algorithmic recommendation systems. Tort law is ill-equipped to price the marginal contribution of one platform's engagement features to a diffuse and multifactorial health trend. And state-by-state litigation creates a patchwork that benefits the incumbents who can afford the lawyers.

The better answer is regulatory, not judicial. Congress should update the legal framework governing social-media platforms in a way that addresses the incentive problem directly. That might include requirements for age assurance that apply across the industry, transparency obligations for how recommendation algorithms work, or standards for the design features that research links to harmful outcomes. It would be more honest than asking state courts to play legislator.

For investors, the immediate financial impact of the New Mexico ruling is negligible. $942 million is less than two weeks of Meta's net income. But the legal overhang is part of a wider pressure on the stock, which is trading at roughly $589, well below its 52-week high of $796 in August 2025. That decline reflects not child-safety penalties but the company's spending spree: capital-expenditure guidance for 2026 has been raised to $125 billion to $145 billion, a reflection of the AI infrastructure build-out that is reshaping the entire technology sector.

The ruling sends a message that companies will be held accountable when their product design knowingly puts users at risk. The trouble is that the message has been sent to a company with $81 billion in the bank and 3.6 billion daily active users. The $567 million fine will be paid. The question is whether the legal system has found a way to change the incentives that produced the harm in the first place. So far it has not.

That is a problem for lawmakers, not judges.