Large language models like ChatGPT become chatbots through a process of reinforcement learning from human feedback: humans rate different outputs for the same prompt, and the model is then trained to prefer the outputs that humans rated as better.

factualpending

Speaker

Melanie Mitchell

Evidence Quote

“a human a army of humans would come and rate outputs to different prompts and say no that's the best one and then you train the model to have this to prefer the same outputs that the humans prefer so that's human feedback”

Source

The Future of Artificial Intelligence— Santa Fe Institute
Created: 8/12/2026, 6:19:21 PM

Your notes

Loading notes...