Grok 4.6 leads in biosecurity refusal without compromising research utility
ns123abc · x · 2026-09-02
SpaceXAI shared an independent evaluation by LatchBio highlighting Grok 4.6's biosecurity capabilities. On the BioSecBench-Refusal benchmark, Grok 4.6 was the only model to score above 50% on both refusing disguised red-team tasks and completing routine dual-use research tasks. The report notes that this safety behavior stems from model intelligence rather than input classifiers, without degrading performance in general biological benchmarks.
More from Models
- Paying Google AI subscriber blocked: custom-domain Workspace accounts can't access AI Studio — ChrisRackauckas · 2026-10-18
- Running Ling 3.1 Flash (560B MoE, 25B Active) Across 4 DGX Sparks via llama.cpp RPC — BerryBlush_Rogers691 · 2026-10-18
- Users press OpenAI on long-delayed Adult Mode a year after it was promised — ___Patrice___ · 2026-10-18
- Polymarket opens betting market on who releases the next Gemini Argon model — Polymarket · 2026-10-18
- Grok matter-of-factly lists which groups user was racist against, exempting Africans and Latin Americans — teortaxesTex · 2026-10-18
- A quick test for benchmaxxed demos: swap in Mirror's Edge and see if it generalizes — teortaxesTex · 2026-10-18