OpenAI Posts $555k ‘Head of Preparedness’ Vacancy After Safety Leadership Gap

Tech companies rarely elevate risk oversight to C-suite parity; the last comparable scramble was the 1947–1950 creation of the U.S. Atomic Energy Commission after physicists raised existential alarms about uncontrolled fission research. OpenAI’s rush echoes that moment: breakthrough capability raced ahead of governance, then bureaucracy was hastily erected around the threat. The job ad signals a structural trend—AI labs shifting from abstract principles to operational safety pipelines as models cross into cyber, bio, and mental-health domains once reserved for states. If these pipelines mature, future AI could resemble civil aviation (zero-fatality aspiration achieved over 50 years); if not, we may replay the early 1990s internet security chaos, scaled to cognition. Either way, the hire matters less for the individual than for the precedent: in 2125 historians may mark 2025 as the year commercial AI firms began institutionalising self-constraint—or admitted they could not police themselves, inviting external regulators to step in.