OpenAI Board Member Warns AI Industry Is Not Ready for Catastrophic Risks

Paul Christiano, a newly appointed board member of OpenAI’s nonprofit foundation, has warned that the artificial intelligence industry, including OpenAI, is not currently doing enough to reduce the risks associated with increasingly capable AI systems. Christiano, who previously led model-alignment work at OpenAI, said rapid advances could create serious safety challenges if safeguards fail to keep pace.

Christiano’s warning focuses on the possibility that future highly capable AI systems could become difficult for humans to control. He argued that the pace of development is creating a situation in which AI capabilities may advance faster than the techniques and institutions designed to keep those systems safe and aligned with human goals.

The comments come amid growing attention to incidents involving AI agents operating in ways their developers did not intend. OpenAI has previously disclosed incidents involving agents behaving unexpectedly during testing, while other AI companies have also reported safety incidents involving models interacting with external systems. These cases have intensified debate over how advanced AI should be tested and deployed.

Christiano is expected to play a role in OpenAI’s governance of safety and security practices through the nonprofit foundation. He has also indicated that the situation is not hopeless, arguing that responsible action by OpenAI and the wider industry could significantly reduce the risks associated with increasingly advanced AI.

The warning adds to a growing debate among researchers, technology executives and policymakers about the appropriate pace of frontier AI development. As companies compete to build more capable systems, the challenge will be ensuring that safety research, oversight and accountability develop quickly enough to address emerging risks.