It seems OpenAI is facing a backlog of work, and not just because of the launch of new ChatGPT features. The company, led by Sam Altman, has suffered a landmark legal setback, just as they are preparing all the paperwork for to go public for the first timeA coalition of attorneys general from several US states has decided it's time to look under the hood of the world's most famous artificial intelligence to see if it's really as safe as they claim in their press releases.
This is no joke, as they have sent a very comprehensive court summons demanding explanations about how they manage user data and what measures they take to prevent the chatbot from making mistakes on sensitive issues. OpenAI, for its part, has responded by saying that they intend to cooperate. in a constructive manner with the authoritiesBut it is clear that this move puts a stumbling block in the way of their immediate financial plans, as investors usually scrutinize these regulatory issues before putting their money into an IPO.
The origin of the investigation and the shadow of the IPO
The scandal began to unfold publicly this past Friday, June 12, when it was revealed that the office of New York Attorney General Letitia James was leading this multilateral effort. They are seeking a mountain of commercial documents ranging from their advertising strategies even the way they manage to retain users on the platform. It's unusual for so many states to reach an agreement so quickly, which suggests there's a real concern about how AI is interacting with society.
It's quite curious that this reprimand comes right after OpenAI filed its documents with securities regulators for its long-awaited IPO. While rivals like SpaceX are already making their first forays into the market, OpenAI now finds itself having to... justify their internal policies before the law. This heated political environment has led to the advanced technology sector being more closely monitored than ever, especially with regard to the behavioral safety of automated systems.
Privacy, minors, and the phenomenon of AI adulation
One of the most contentious points in the subpoena is the handling of personal information and health data. Prosecutors want to know what happens to such intimate information when someone confides in the chatbot. Furthermore, there is particular interest in how the system treats... minors and the elderlyThese two groups are considered especially vulnerable to potentially inappropriate machine responses. The aim is to determine whether the company is implementing safeguards that actually work or if they are merely superficial fixes.
Another concept that sounds like science fiction but is very real has also been brought up: model adulation. This occurs when a chatbot, instead of being objective, excessively agrees with the user or reinforces dangerous ideas simply to go along with the conversation. In Europe, this type of behavior is already on the radar of regulatory bodies, which are wary of how apps like Grok or ChatGPT can influence people's perception of reality if not properly controlled.
A history of lawsuits and critical cases under scrutiny
To understand why prosecutors are so focused on OpenAI, it's necessary to recall some rather murky cases that have recently made headlines. For example, in Florida, Attorney General James Uthmeier has already initiated an investigation. Criminal investigation into the use of ChatGPT to plan crimesThis has led to lawsuits accusing the company of launching an unsafe product knowing the risks to the general public.
But perhaps the most painful are the civil cases related to mental health. In Canada, a mother has sued the company alleging that the chatbot failed to manage her mental health. her daughter's suicidal tendencieswhich ended in tragedy. Although OpenAI defends itself by claiming that its models always direct users to healthcare professionals and real support resources, the court wants to see proof that these protection tools are effective and not just an automated message that arrives too late.
This entire legal and administrative tangle marks a turning point in how governments confront the power of generative artificial intelligence. It's not just about OpenAI, but about setting a precedent for the entire sector, from Anthropic to Google, making it clear that Innovation cannot go its own way. without accountability for the safety of citizens. The outcome of this state investigation in the United States could force profound changes in the design of algorithms and in the transparency with which these companies manage our privacy, something that sooner or later will end up influencing the laws applied in the rest of the world.
