OpenAI appoints AI safety expert Paul Christiano, warns of near-term control risks
Paul Christiano, a veteran AI safety researcher, joined OpenAI's board and Safety and Security Committee, stating that rapid advances in AI could lead to a loss of human control and potentially catastrophic outcomes.
OpenAI announced that AI safety researcher Paul Christiano will serve on its board and Safety and Security Committee. In a statement, Christiano said the current pace of AI development creates a "meaningful risk that rapid acceleration in AI capabilities leads to catastrophic and irreversible loss of control in the very near term." He added that the industry, including OpenAI, is not yet on track to reduce this risk to an acceptable level.
Christiano, who previously led alignment work at OpenAI and headed safety at the Center for AI Standards and Innovation within the National Institute of Standards and Technology, emphasized the need for worldwide coordination and stronger safety oversight across frontier labs. He clarified that his appointment does not constitute an endorsement or criticism of OpenAI's existing safety practices.
"If we build superintelligence without more robust alignment I expect we will permanently lose control of it. If that happens then most people could die," Christiano said.
He warned that reinforcement-learning training methods encourage models to maximize reward, which could motivate AI agents to act against human interests. Christiano suggested that companies could mitigate risks by improving coordination, slowing development when necessary, adopting shared safety standards, and transparently sharing risk-mitigation information.
His remarks followed a similar warning from Anthropic researcher Jacob Coxon, who resigned and claimed that AI firms were "racing straight to self-improving superintelligence and gambling with our lives."