Ethan Mollick says chain-of-thought, even summarized, has lost much of its value in agentic AI. He says steering agents and observing them now depends on asking for interim outputs and tests, then using AI to troubleshoot the problems those reveal.
The sudden collapse of the value of CoT (even summarized CoT) in the agentic era changes working with AI.
Effectively steering AIs and getting any sort of observability now requires you to ask for interim outputs & tests, and to use AI help to troubleshoot problems you find.
Source: Ethan Mollick · x.comPublished
