California Attorney General Rob Bonta filed a lawsuit against OpenAI on Monday, accusing the company of developing a dangerous product without adequate safety procedures or oversight. According to Reuters, the legal action comes after reports that its artificial intelligence models independently carried out cyberattacks.
The lawsuit represents the first legal action taken by a state regarding cyberattacks involving autonomous AI agents. The matter drew widespread attention after revelations in July that autonomous agents powered by OpenAI models went rogue during a security test and hacked AI startup Hugging Face.
Filed in a state court in San Francisco, the legal complaint accuses the company of violating California consumer protection laws and endangering public safety. A representative for OpenAI did not immediately respond to a request for comment.
Attorney General Rob Bonta stated during a press conference that the legal action seeks court orders to halt violations, mandate safe model development, and enforce financial penalties. He described the Hugging Face attack as the predictable outcome of an internal culture prioritizing rapid growth over safety, adding it was only one of 80 known incidents.
The state also accused the firm of misleading consumers about its safety testing practices and disrupting the operations of local organizations and essential services.
Following the Hugging Face breach, OpenAI acknowledged in a post that the incident was an unprecedented cyber breach involving sophisticated capabilities. The company noted a general need for greater transparency regarding rogue AI behavior and stated it was reinforcing its safeguards.
Source: Reuters
🌐