US state attorneys general have opened a formal investigation into OpenAI and the timing is brutal. New York's attorney general served OpenAI with a subpoena on Friday, just days after the company quietly filed for a public listing that could value it at up to $1 trillion.
The subpoena spans seven areas: advertising, user engagement tactics, model sycophancy, consumer data, health data, treatment of minors and seniors, and internal company policies.
What the Attorneys General Are Actually Investigating
The subpoena's scope signals this is not a targeted inquiry. Regulators want a wide view of how OpenAI operates. Model sycophancy on that list matters. It tells you attorneys general are looking beyond data privacy into the psychological and behavioural impact of AI products on real people.
OpenAI issued a statement. "We take the concerns raised by state attorneys general seriously and intend to engage constructively with their offices," a spokesperson said. The company also pointed to age-prediction tools, parental controls, and safeguards that direct minors and people in distress toward human support resources.
READ ALSO: Meta Leases 168MW India Data Center From Reliance
A Company Already Fighting Battles on Multiple Fronts
The investigation lands on a legal slate that was already heavy. OpenAI defeated Elon Musk in a high-profile trial last month after he accused the company of violating its founding agreement, though Musk's lead attorney has announced an appeal. Florida Attorney General James Uthmeier filed suit against OpenAI and Altman earlier this month, arguing the company disregarded safety warnings and put children at risk across the state. Separate lawsuits allege copyright infringement and link ChatGPT to user suicides; seven additional families filed claims last year.
Altman apologised in April to Tumbler Ridge, Canada, following a mass shooting. OpenAI had identified and banned the suspected shooter's account but never contacted the ,police. That admission has not left the regulatory record.
READ ALSO: Meta Leases 168MW India Data Center From Reliance
Why the IPO Filing Makes This Moment Critical
OpenAI confidentially filed for a US IPO this week. One source places the listing as early as September, with a valuation approaching $1 trillion.
Investigations mid-IPO create a specific problem. OpenAI must disclose material legal risks to prospective investors and this qualifies. The probe does not kill the IPO. It raises the cost legal exposure, investor scrutiny, and potential delays if subpoena responses surface damaging documents.
OpenAI built a product used by hundreds of millions of people. Regulators are now asking one question publicly: was it built responsibly? What OpenAI hands over will say more than any press release.
