factual

LLMs Excel at Interpolation Not Extrapolation

Large language models are very good at interpolating within the space of everything humans have done and said—which can seem highly creative—but extrapolating to genuinely brand-new places is much harder, so they should be understood as performing creative remixing rather than truly innovative extrapolation.

factualpending

Speaker

Yejin Choi

Evidence Quote

“it should do more of the interpolation than true innovative extrapolation”

Source

248 | Yejin Choi on AI and Common Sense— Sean Carroll's Mindscape
Created: 6/13/2026, 12:23:11 AM

Your notes

Loading notes...