Security veteran mocks AI safety folks for rediscovering 50-year-old threat classes

inductionheads · x · 2026-09-19

Security researcher @0x4D31 quips that AI safety/alignment people are rediscovering threat classes cybersecurity has dealt with for over 50 years, then going on TV as if they've discovered a new species of risk — a pointed critique of AI safety's tendency to relabel classic security problems as novel.

Original post →

More from AGI Musings

AGI Musings channel →