Stephen Witt raises AI safety alarm in New York Times essay: 'It's bad. Really bad.'

MariusHobbhahn · x · 2026-09-19

Author Stephen Witt has published an essay in the New York Times sounding the alarm on AI safety, teasing his central conclusion bluntly: the state of AI safety is "bad. Really bad." The piece adds another prominent insider voice to mainstream coverage of the gap between rapidly improving frontier models and inadequate safety measures.

Original post →

More from AGI Musings

AGI Musings channel →