Computerphile: The AI language we can't read — Neuralese ft. Rob Miles
Computerphile · youtube · 2026-09-10
Computerphile sits down with AI safety advocate Rob Miles to explore what happens if LLMs stop reasoning in human-readable chain of thought and start communicating in a latent language only they understand. The video unpacks the idea of 'neuralese', why CoT oversight could break down, and what this means for AI safety and interpretability.
More from Research
- AgentGrad: intervention-guided prompt optimization hits SOTA with 2.5x faster tuning — _akhaliq · 2026-09-10
- David Chalmers talks Anthropic's j-space and global workspace theory aboard a boat in the Galapagos — PeterBowdenLive · 2026-09-10
- A visual deep-dive catalogs 42+ representations of 3D, praised by HF engineer — pcuenq · 2026-09-10
- Ben Recht's forecasting lecture argues probability conflates frequency and belief — beenwrekt · 2026-09-10
- Spiced self-play accepted at CoRL: just 30 minutes of human data biases agents to right conventions — EugeneVinitsky · 2026-09-10
- Understanding FlashAttention: A Handbook Tracing FA1 to FA4 and Why HBM Traffic, Not FLOPs, Is the Bottleneck — techNmak · 2026-09-10