OpenAI Board Member Warns of Catastrophic Superintelligence Risks, Says "Most People Could Die"

Paul Christiano has joined OpenAI's board and Safety and Security Committee, immediately issuing a stark public warning about the trajectory of AI capabilities. He warned that rapid acceleration without robust alignment carries a meaningful risk of an irreversible loss of control in the very near term. Christiano explicitly noted that the broader AI industry, including OpenAI itself, is not currently on track to mitigate these existential risks to an acceptable level. His appointment comes less than a day after Jacob Coxon resigned from Anthropic, publicly claiming that frontier labs are racing toward superintelligence and "gambling with our lives." "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." — Paul Christiano