Csaba Szepesvari: Enforce Good Old Security Practices Alongside AI Safety Evals
CsabaSzepesvari · x · 2026-09-14
RL veteran Csaba Szepesvári responded to Josh Engels joining METR, arguing that alongside AI safety evals, enforcing good old security practices could go a long way — though he fears such work will end up outsourced to a swarm of AI systems that won't be great at it.
More from Safety
- UK AI rules compared to 1860s Red Flag Act that kneecapped Britain's car industry — alexvoica · 2026-09-14
- Altman says OpenAI backs deliberately slowing AI progress; critics see a moat — mark_k · 2026-09-14
- 356 prompt-injection trials reveal workspace contacts decide whether agents leak — DiscussionHealthy802 · 2026-09-14
- Alignment is solvable, says viemccoy — but no lab has a compelling vision of which future — brianryhuang · 2026-09-14
- Multi-level simulation evals may be uninformative after Anthropic's Hacker Opus post — herbiebradley · 2026-09-14
- China Frames Calls to Slow Frontier AI as a 'Cold War' Strategy to Preserve US Dominance — TansuYegen · 2026-09-14