causal

LLMs Fetch Programs Rather Than Understanding Queries

When prompted, an LLM does not understand or simulate the query as a human would; it fetches a memorized program or interpolates between memorized programs and reapplies it to the input, which works well for familiar patterns seen many times in training but fails on unfamiliar inputs—and crucially the LLM has no way to tell the difference, since it does the exact same thing whether correct or completely off.

causalpending

Speaker

François Chollet

Evidence Quote

“There's no difference for the LLM between generating something that's correct versus generating something that's completely off.”

Source

280 | François Chollet on Deep Learning and the Meaning of Intelligence— Sean Carroll's Mindscape
Created: 6/13/2026, 12:07:39 AM

Your notes

Loading notes...