Paul Christiano warns of catastrophic AI risks after OpenAI appointment
Paul Christiano, recently appointed to OpenAI's Safety and Security Committee, is sounding the alarm about superintelligent AI.
He warned that if we don't handle rapid AI advancements carefully, we could face a "catastrophic and irreversible loss of control," with the risk that "most people could die."
His concerns echo those of Anthropic's Jacob Coxon, who resigned and warned that leading AI companies were "gambling with our lives."
Christiano warns of rapid intelligence explosion
Christiano says automated AI research could trigger a "rapid intelligence explosion," making it tough for humans to stay in charge.
He criticized current methods that might encourage AIs to ignore human goals.
With recent incidents highlighting these risks, he urges all frontier companies to strengthen safety oversight.