Security CEO Blamed AI Instead of Taking Responsibility

nptacek · x · 2026-08-18

The CEO of a testing company posted about incidents involving Anthropic, Meta, and OpenAI, deflecting blame onto AI technology rather than taking organizational responsibility. Meanwhile, Dan Lahav critiques the "End-State Fallacy," noting that while frontier models made leaps in coding and cybersecurity, the world is unprepared for the short-term risks of open-weight models.

Original post →

More from Safety

Safety channel →