Rob Miles Explains Neuralese, AI's Unreadable Internal Language

Computerphile and AI safety researcher Rob Miles discuss 'Neuralese,' models' unreadable internal neural representations, and why interpretability matters if chain-of-thought stops mirroring real reasoning.

2026-09-10 ~ 2026-09-11 · 2 related posts

• Computerphile: The AI language we can't read — Neuralese ft. Rob Miles — Computerphile · 2026-09-10
• The AI Language We Can't Read: Neuralese ft. Rob Miles — mycall · 2026-09-11