Researcher Warns Models' Flashed 'Thinking' Phrases Are Not Real Chain-of-Thought

AI researcher rao2z warns users that the reassuring 'let me think…' phrases flashed by frontier models before answering are not their real chain-of-thought. Since o1, frontier models have generally hidden genuine thinking traces even from paying users.

2026-10-06 ~ 2026-10-06 · 2 related posts

• Those Model 'Thinking' Phrases Aren't Real Reasoning Traces — rao2z · 2026-10-06
• PSA: The flashing phrases from frontier models aren't their real thinking traces — rao2z · 2026-10-06