OpenAI is facing a joint investigation by attorneys general from multiple U.S. states. New York Attorney General Letitia James has issued a subpoena to the company, with the probe covering issues such as user data privacy protection, safety measures for minors, potential advertising business, and whether AI models have a tendency to "ingratiate" themselves with users.
According to sources familiar with the matter, the investigation involves at least five states, with investigators focusing on how OpenAI collects, stores, and uses user conversation data, and whether products like ChatGPT have taken adequate measures to prevent minors from accessing harmful content. In addition, regulators are also paying attention to whether OpenAI plans to monetize through advertising, and whether the model training process involves designs that overly cater to user preferences.
OpenAI is currently in the IPO preparation stage, with the company valued at over $100 billion. If the investigation finds violations, it could lead to fines, business adjustments, or even delays in the listing timeline. Over the past year, OpenAI has repeatedly adjusted its product strategy in response to regulatory concerns.
Analysts point out that this multi-state investigation reflects that local levels in the U.S. are accelerating efforts to fill the gap in federal AI regulation. The European Union's Artificial Intelligence Act came into effect this year, and some U.S. states hope to protect consumer rights through similar actions. OpenAI CEO Sam Altman has repeatedly stated that the company welcomes reasonable regulation, but also emphasized that excessive restrictions could hinder innovation.
From an impact perspective, the investigation results may force OpenAI to further strengthen data anonymization and introduce stricter age verification mechanisms. Regarding the advertising business, OpenAI may need to clearly define the boundaries of data usage between free and paid users to avoid privacy infringement allegations.
On the issue of model "sycophancy," investigators are concerned that AI may generate misleading information to cater to users, potentially amplifying the risk of spreading false information. OpenAI has committed to improving model alignment techniques, but the specific results still need time to verify.
Overall, this regulatory action marks the AI industry entering an era of strong regulation. OpenAI and other leading companies need to find a balance between technological innovation and compliance; otherwise, the IPO path will be fraught with uncertainty.
© 2026 Winzheng.com 赢政天下 | 转载请注明来源并附原文链接