Should AI models pass NIST-style safety checks before shipping? Industry voices say it's simple
aronchick · x · 2026-09-15
- Peter Yared argues AI safety isn't complicated: models should pass a transparent set of controls before shipping, benchmarked to government standards like NIST Cybersecurity or industry standards like ISO 27001. Like all software, it only adds an evaluation-and-remediation step without slowing development.
- He flags one interesting shift: secure SDLC currently requires human review of every change, and AI might soon be trusted to do that review — but a model should never review itself.
- aronchick amplified the take, calling it common sense and questioning why so many people oppose it.
More from AGI Musings
- Solo dev argues LLMs shouldn't be the default center of an agent runtime — HmmmThisIsOdd · 2026-09-15
- "Models are slowing down" narrative disputed, as Sabine Hossenfelder argues labs slow AI to dodge lawsuits — zetalyrae · 2026-09-15
- Oxford's Carissa Véliz: Better Science, Not AI Slop Tainting Academic Literature — CarissaVeliz · 2026-09-15
- Jack Clark: stories about AI literally become training data for the next models — jackclarkSF · 2026-09-15
- Jensen Huang revisits the AI predictions that never came true at All-In Summit 2026 — tetsuoai · 2026-09-15
- Critics say "Responsible Scaling Policies" is now "Pacing the Frontier" — same old evals — DavidSKrueger · 2026-09-15