normative

Supervised Fine-Tuning on Transcripts Is a Bad Target

Naively fine-tuning a base model to predict every token observed during a session is a nonsensical learning target, because people improve at jobs by consolidating a few relevant insights rather than recalling every transcript with perfect fidelity.

normativepending

Speaker

Dwarkesh Patel

Evidence Quote

“the way you get better at your job is not by recalling the transcript of what happened through every single day with perfect fidelity.”

Source

Dwarkesh Patel Podcast - The next big breakthrough will be AIs learning on the job— Dwarkesh Patel Podcast
Created: 11/17/2026, 6:25:50 AM

Your notes

Loading notes...