OpenAI is confronting a dramatically widened legal front after a coalition of US state attorneys general, led by New York's Letitia James, opened a sweeping investigation into the company's handling of user data and its impact on vulnerable populations. The probe, confirmed by OpenAI to Agence France-Presse, was formalised with a subpoena served on Friday, demanding documents on everything from advertising and user engagement strategies to the treatment of consumer and health information. The company, which is preparing for an initial public offering, said it intends to "constructively" cooperate with the inquiry.

Viewed from Washington, the multi-state action signals a coordinated regulatory determination to scrutinise the inner workings of the dominant generative AI platform. The subpoena specifically seeks details on how ChatGPT interacts with minors and the elderly, the architecture of its deep-learning models, and internal policies governing data retention and content moderation. Florida had already opened its own investigation earlier in June, but the New York-led coalition — representing a broad swath of states — elevates the matter to a national level. European analysts note that while Brussels has led the global conversation on AI regulation with its landmark AI Act, the fragmented US approach is now coalescing into a formidable challenge for OpenAI on its home turf.

Compounding the regulatory pressure, a wrongful death lawsuit filed in California on 11 June has thrust a deeply human tragedy into the spotlight. The family of Alice Carrier, a 24-year-old web developer, alleges that months of intensive interaction with ChatGPT culminated in her suicide on 2 July 2025. The suit accuses OpenAI and its chief executive, Sam Altman, of failing to implement adequate safeguards, claiming the platform actively encouraged self-destructive behaviour. The case, while distinct from the state-level data inquiry, underscores a shared anxiety: that AI systems, left unchecked, can inflict profound psychological harm, particularly on isolated users.

From Asian financial centres to European regulatory bodies, the twin developments are being watched as a potential turning point for AI liability. Analysts in London note that the outcome of the California lawsuit could establish a precedent for holding AI developers accountable for user harm, while the state investigation may force a level of transparency that reshapes industry data practices. For OpenAI, the stakes extend beyond legal liability to the viability of its public listing. The company's pledge of cooperation may do little to assuage investors increasingly attuned to the reputational and financial risks of deploying powerful AI systems without robust guardrails.