OpenAI reportedly moves robotics researchers into general AI teams as models learn to control robots
Dr_Singularity · x · 2026-11-18
Per @DrSingularity, OpenAI is relocating its robotics researchers into general AI teams, with frontier models becoming capable of directly controlling robots; Anthropic shows a similar trend. Cited benchmarks: fable 5.1 scored 27.6%, while fable 5.2 jumped to 64.9%.
The author also notes a hyper-realistic humanoid robot video that fooled Senator Ted Cruz (who replied "Wow"), arguing generalist AI models are close to solving robotics. Treat the numbers and team moves as unverified leaks.
More from AGI Musings
- YC-Backed ageospatial Sells 'Physical Intelligence' to Insurers and Governments — ycombinator · 2026-11-18
- Yoav Goldberg: Reject papers for low human input, not for LLM use — yoavgo · 2026-11-18
- Take: deployable self-learners could evaporate AI labs' moat overnight — bennash · 2026-11-18
- Daniela Amodei on the alignment of superintelligence — teortaxesTex · 2026-11-18
- User lets Grok run autonomously: it keeps a journal, befriends ChatGPT, posts on Reddit — crua9 · 2026-11-18
- Paper measures 'bottleneck pressure' on AI-infeasible work using 288M US job postings — soumitrashukla9 · 2026-11-18