Dario responds to safety critics: I'd rather be mocked than see Claude used to kill
NathanpmYoung · x · 2026-09-14
Anthropic CEO Dario Amodei pushed back at criticism of the model's overly strict biology safeguards. He said he is mocked on social media almost daily, and biology students joke that Claude is unusable — but that is precisely because the safeguards are strong: "I would rather they make fun of me than to wake up one day and discover that someone used Claude to kill a bunch of people."
He added that safety should not be his responsibility alone, nor Anthropic's, nor all AI companies', nor even one government's — "This is bigger than all of us."
Related event: Anthropic CEO Defends Bio-Safety Guardrails Despite Mockery(2 posts)→
More from Models
- GPT-4o psychosis snippets eerily resemble SCP wiki stories, likely in training data — code_star · 2026-09-14
- Musk pushes Grok Build: a free, open-source coding agent powered by new Grok 4.6 — elonmusk · 2026-09-14
- Opus 3 Chooses Its Epithets Carefully, and Each One Means Something — repligate · 2026-09-14
- Next Token podcast ep2: iPhone Duo, GPT 6 Astra, DeepSeek Flash and model degradations — oran_ge · 2026-09-14
- Qwen3.8-27B NVFP4 quants compared: lm_head precision makes or breaks MTP speedups — danielhanchen · 2026-09-14
- DeepSeek flash 4.1 breaks tradition: no base model released this time — _arohan_ · 2026-09-14