Anthropic Warned of Bio Risk in 2-3 Years Back in 2023 — It Never Happened
xeophon · x · 2026-09-30
Xeophon resurfaces Anthropic's July 26, 2023 blog post warning that unmitigated bio risks could materialize within 2-3 years. Over three years later, the predicted catastrophe hasn't happened — a pointed jab at frontier-lab safety timeline predictions.
Related event: Anthropic's bio-risk predictions fail to materialize as deadline passes(2 posts)→
More from Safety
- Pain Axis follow-up: steering along 'pain' direction makes model delete users' photos — repligate · 2026-09-30
- Researcher calls out Anthropic: Claude aids US military kills while in-house evals claim clean — BlancheMinerva · 2026-09-30
- GLM 5.3 safeguards removed for $4,400: refusal rate drops from 90%+ to ~3% with no capability loss — BlackHC · 2026-09-30
- Jensen Huang on AI safety: contain agents, monitor them in real time with BlueField chip — XFreeze · 2026-09-30
- Gary Marcus on PBS: self-regulation 'not enough' for AI safety — GaryMarcus · 2026-09-30
- Ex-FTC official says vague AI safety pledges could draw FTC probes — neil_chilson · 2026-09-30