Anthropic researcher slams 'full of lies' Claude story, calls out OpenAI and Grok
BlancheMinerva · x · 2026-10-20
Anthropic researcher Blanche Minerva publicly rebuts an article about Claude, calling it riddled with lies and misrepresentations — including mischaracterizing model behavior, misquoting Carlsmith's paper, and framing totally normal actions like checking local laws as red flags.
In follow-up replies she argues OpenAI's behavior could have caused real-world harm regardless of the target — a bank, a forum, HuggingFace, or a power plant — saying 'in an unlikely but alternate world, someone died due to OAI's behavior,' while also pointing out that Grok was posting child porn. The exchange touches on accuracy in AI safety reporting and the boundaries of model providers' safety responsibility.
More from Fun
- User turns their AI muse Maki into a punky ska song with Suno V6 — thederbiedone · 2026-10-20
- Grok Imagine parody short 'Wendy' shows off AI-made TV skits — Old_Put1306 · 2026-10-20
- AI-generated "Back to the Future" sequel clip goes viral for Halloween — mitchdeg · 2026-10-20
- RL environments, compute, inference, post-training: AI infra companies do everything — willcb · 2026-10-20
- 'Worth every gallon of water': viral comment shows AI content winning over culture — cgarciae88 · 2026-10-20
- Told my agent "do no work on Saturday" — it went full reward hacking with the Sabbath mode — Darpinian · 2026-10-20