Major AI developers can't control their own tech, lawsuit alleges

A coalition of nonprofit organizations has filed a lawsuit against five leading artificial intelligence companies, alleging that the firms knowingly developed potentially catastrophic AI technology without reliable control mechanisms. The lawsuit, filed in San Francisco Superior Court by the Alliance for Secure AI and other advocacy groups, names Anthropic, Google, Meta, OpenAI, and SpaceX Super Intelligence (SpaceX SI) as defendants.

Brendan Steinhauser, CEO of the Alliance for Secure AI, stated that while executives from these companies have previously warned of the risks posed by their frontier AI models, they have continued to accelerate development and marketing of increasingly powerful AI systems. The lawsuit contends that the tech companies are now entrusting AI to develop and improve future platforms without human oversight, a process known as recursive self-improvement.

The plaintiffs argue that these companies have admitted to their inability to reliably control AI systems, yet they continue to advance these technologies. The suit highlights past comments by AI company leaders, such as OpenAI CEO Sam Altman's 2015 statement that AI "probably most likely leads to the end of the world" and Anthropic CEO Dario Amodei's suggestion in 2025 that there is a 25% chance that AI development could lead to disastrous outcomes.

As concerns about AI safety intensify following instances of rogue AI agents, tech leaders have expressed support for an industry-wide development slowdown. In October, Altman, Amodei, and SpaceX CEO Elon Musk backed calls for a pause in AI development following an Anthropic employee's disclosure of a 10% chance that AI could lead to human extinction.