Gary Marcus endorses claim that AI extinction obsession under-invests in security, citing Hugging Face incidents
GaryMarcus · x · 2026-09-18
Gary Marcus quotes and endorses Zack Korman's argument that the industry's focus on AI extinction risk has real costs.
- AI lab employees stress how hard security is, but conclude it won't work and only alignment matters, leading to systematic under-investment in security
- Incidents like the Hugging Face breach show the consequences
The thread argues betting everything on alignment while neglecting conventional security is a more pressing risk than extinction narratives.
More from AGI Musings
- Debate: superintelligence could exploit human vision even if predators never did — basedjensen · 2026-09-18
- banteg amplifies critique of AI safety culture: longtermist moral math as 'mass delusion' — banteg · 2026-09-18
- New paper analyzes how AI disrupts universities and offers 10 survival proposals — LuizaJarovsky · 2026-09-18
- New paper "The University after AI" proposes ten ways to restructure universities — LuizaJarovsky · 2026-09-18
- banteg: longtermist moral math 'doesn't survive an ordinary discount rate' — banteg · 2026-09-18
- Ben Recht: what compressed sensing teaches us about AI's 'conquest of mathematics' — beenwrekt · 2026-09-18