Meta finds itself in the crosshairs of a major legal battle over youth safety, with four American states pursuing billions of dollars in damages and demanding sweeping operational changes to Instagram and Facebook. The company's legal representatives have mounted a vigorous defence, arguing that Meta takes children's mental health seriously and has demonstrated genuine commitment to child protection through a steady stream of safety initiatives. Yet this courtroom conflict exposes a fundamental disconnect: while Meta points to years of incremental changes, critics contend that beneath the surface, these reforms amount to little more than public relations exercises designed to deflect pressure rather than meaningfully protect young users.
The crux of the dispute centres on Meta's alleged complicity in fuelling a youth mental health crisis through deliberately engineered platform features designed to capture and retain children's attention. This framing represents a significant evolution in how regulators and advocacy groups are approaching tech accountability—moving beyond concerns about content moderation to challenge the very architecture of engagement mechanisms themselves. Paul Schmidt, Meta's lawyer, acknowledged in opening statements that the company recognises the importance of children's wellbeing but emphasised that Meta has acted responsibly within industry norms. However, critics including Marc Berkman, executive director at the Organization for Social Media Safety, counter that Meta's institutional commitment to change has always been reactive rather than proactive, emerging only when public outcry becomes too loud to ignore.
Examining Meta's actual safety infrastructure reveals a patchwork of protective tools primarily focused on managing harassment, offensive content, and illicit contact between adults and minors. Instagram users can block other accounts, report problematic material, and filter offensive comments—features that have accumulated gradually since the platform's inception. The company introduced tools in 2016 allowing users to delete unwanted comments and subsequently added warnings encouraging users to reconsider potentially harmful posts before publication. More recently, Meta implemented controls preventing unconnected adults from initiating private conversations with teenagers and heightened age verification procedures, though Meta concedes that some users still circumvent these restrictions by providing false birthdates during registration.
Meta has also positioned itself as addressing sensitive mental health content by placing resource warnings and support information prominently within Instagram search results related to suicide, self-harm, and eating disorders. The company's flagship Teen Accounts feature, launched in 2024, operates as a walled garden with content restrictions mirroring PG-13 movie standards unless parental authorisation permits otherwise. In theory, these measures demonstrate institutional acknowledgment of youth vulnerability. Yet this defensive posture obscures a more troubling reality: advocates argue that Meta introduced most protections only after sustained external pressure, and questions persist about whether implementation matches the rhetoric surrounding these safety announcements.
A September 2025 report from former Meta employee Arturo Béjar and four online safety advocacy nonprofits delivered a damaging verdict on Meta's claims. Examining 47 of the company's 53 teen safety features on Instagram, the researchers discovered that approximately 60 percent were either unavailable or functioned differently than advertised. Haley Hinkle, policy counsel at Fairplay, one of the report's co-authors, characterised Meta's Teen Accounts initiative as heavily marketed but substantially lacking in protective substance. Meta dismissed these findings as misleading and dangerously speculative, but the report's detailed methodology and specific audit findings suggest the company's defensive posture reflects institutional defensiveness rather than confident vindication. The discrepancy between what Meta publicises and what researchers empirically find on platforms raises uncomfortable questions about whether the company's safety narrative is deliberately constructed for external consumption while internal design priorities remain unchanged.
The most academically substantiated criticism targets Meta's failure to address what researchers call attention-capture deceptive design features—the addictive mechanics engineered into platform architecture itself. Ashley Shea, a Cornell University PhD candidate specialising in social media design, identifies infinite scrolling, algorithmic recommendations designed like casino reward systems, push notifications, and mechanised content-refresh pulls as cognitive traps that exploit developmental vulnerabilities in adolescent brains. These features deliberately prey on young people's neurological need for reward and social validation, creating compulsive usage patterns that fundamentally differ from traditional content moderation issues. While Shea acknowledges Meta's genuine efforts to combat cyberbullying and sexual predation, she identifies a conspicuous absence of equivalent institutional energy devoted to mitigating these addictive design mechanisms—a gap that suggests priorities remain commercially driven rather than safety-focused.
Meta's time management toolkit—comprising parental supervision controls and in-app nudges alerting teenagers to excessive late-night usage or repetitive content consumption—ostensibly addresses addiction concerns. Schmidt referenced these features as evidence of Meta's evolving approach, yet advocates highlight their fundamental limitations. The so-called Take a Break feature, according to Béjar's testimony, requires individual users to activate the tool themselves, and empirical evidence indicates very few actually do so. Hinkle characterises all time management protections as entirely voluntary and inadequately designed, describing them as coming far too late in the company's trajectory to constitute genuine institutional commitment. This pattern—offering tools that place burden entirely on users rather than redesigning platforms themselves—reflects a consistent corporate strategy of appearing responsive while maintaining fundamentally unchanged business models.
Marc Berkman articulates the structural conflict underpinning this entire debate with particular clarity: technology companies exist to maximise revenue, and across social media platforms, that economic imperative translates directly into maximising children's engagement duration. From this perspective, safety initiatives become acceptable only insofar as they maintain public legitimacy without disrupting the engagement mechanics that generate advertising revenue. Legal challenges now emerging differ meaningfully from earlier content-focused complaints; they characterise Meta's platforms as defective products by design, not merely by the content they host. This reframing shifts the argument from debatable questions about whether specific posts cause harm to more fundamental questions about whether the platforms' underlying architecture is inherently predatory toward young users.
The litigation emerging from California represents a potential watershed moment for tech accountability because it challenges not individual harmful content but the engineered systems that exploit developmental psychology to maximise user dependence. Research indicates that mounting legal and political pressure creates economic incentives for companies to adapt their business models, and momentum continues building as legislators worldwide examine comparable regulatory approaches. Ashley Shea notes that for-profit companies ultimately respond to economic pressure, and continued litigation may prove more effective at forcing genuine platform redesign than voluntary corporate initiatives. The trajectory suggests that Meta and comparable tech giants will face increasingly difficult choices between maintaining profitable engagement-maximisation strategies and genuinely restructuring platforms to protect developing brains from deliberately engineered addiction mechanisms.
