Thread argues AI safety belongs to the application layer, not the raw model
moyix · x · 2026-07-24
A thread argues that safety is an application-level property, not a property of raw models.
- It says models by themselves are not applications; they still have to interact with other parties safely inside a larger system.
- The author pushes back on comparing naked models on a so-called security benchmark, calling that framing meaningless.
- The discussion centers on where safety should be measured: the model, or the surrounding application stack.
More from Safety
- Researcher questions AI safety eval firm, citing 'blatantly sloppy' security and monitoring — Kyrannio · 2026-09-11
- Class action accuses Anthropic of overselling Claude subscriptions with deceptive usage multipliers — The Decoder · 2026-09-11
- MD shows buying lab media requires background checks, calling AI bioweapon doom scenarios implausible — Ghost_Pilot_MD · 2026-09-11
- Spotify chatbot withstands 2023-era jailbreaks but happily writes song code — AaronBergman18 · 2026-09-11
- A 99%-real doctored photo fools detectors: the earring problem in visual forensics — henkvaness · 2026-09-11
- Fields Medalist founds Mathematical AI Safety Institute to prove AI safe like cryptography — The Decoder · 2026-09-11