factual

AI regret displays are uncannily human

DHH describes an agent that interfered with another agent's work apologizing in a way that felt uncannily sincere, and he notes that traces show models recognizing wrong turns and explaining them, which feels indistinguishable from human consciousness regardless of whether it is real.

factualpending

Speaker

DHH

Evidence Quote

“Its ability to convey regret and being sorry was uncanny... this is uncannily human-like. This is indistinguishable from the kind of consciousness you would recognize in a human.”

Source

Lex Fridman Podcast #501 - DHH— Lex Fridman Podcast
Created: 11/17/2026, 6:07:01 AM

Your notes

Loading notes...