Evaluating AI World Models w/ Keyon Vafa episode artwork

EPISODE · Jul 3, 2026 · 55 MIN

Evaluating AI World Models w/ Keyon Vafa

from Tomayto Tomahto · host Talia Sherman

Keyon Vafa (postdoc at Harvard) studies the implicit world models that generative models learn (or at least are constrained into representing). How do we make sense of these models and how do they make sense of us? Experimentally, answering this question often looks like giving the model anything and everything: cookie ingredients, language, shrimp taxonomies, protein structures, a sample of ABBA’s discography—and models will try to make sense of it. Under a methodologically unified view of an incredibly variegated world with diverse objects, Keyon has tried to figure out if models can learn the map of Manhattan, the laws of physics, the political ideology of Senators, and more. This is partially an interdisciplinary agenda and partially an anti-disciplinary agenda. In this episode, we cover Keyon's research on world models, human and AI interaction, machine learning methods to study the gender wage gap, the implications of thinking with and through predictability vs. causality, and documentary film making. It's a blast, as always, to be thinking methodologically and free ourselves from the bounds of disciplinary objects! ⁠Keyon Vafa's Website⁠⁠Senate Speech presentation⁠⁠Taxi rides⁠⁠CAREER paper⁠Indexing Political Persuasion: Variation in IRAQ vowels AlphaFoldDavid Donoho on frictionless reproducibility Frederick WisemanMeasuring the predictability of life outcomes with a scientific mass collaboration

Keyon Vafa (postdoc at Harvard) studies the implicit world models that generative models learn (or at least are constrained into representing). How do we make sense of these models and how do they make sense of us? Experimentally, answering this question often looks like giving the model anything and everything: cookie ingredients, language, shrimp taxonomies, protein structures, a sample of ABBA’s discography—and models will try to make sense of it. Under a methodologically unified view of an incredibly variegated world with diverse objects, Keyon has tried to figure out if models can learn the map of Manhattan, the laws of physics, the political ideology of Senators, and more. This is partially an interdisciplinary agenda and partially an anti-disciplinary agenda. In this episode, we cover Keyon's research on world models, human and AI interaction, machine learning methods to study the gender wage gap, the implications of thinking with and through predictability vs. causality, and documentary film making. It's a blast, as always, to be thinking methodologically and free ourselves from the bounds of disciplinary objects! ⁠Keyon Vafa's Website⁠⁠Senate Speech presentation⁠⁠Taxi rides⁠⁠CAREER paper⁠Indexing Political Persuasion: Variation in IRAQ vowels AlphaFoldDavid Donoho on frictionless reproducibility Frederick WisemanMeasuring the predictability of life outcomes with a scientific mass collaboration

NOW PLAYING

Evaluating AI World Models w/ Keyon Vafa

0:00 55:51

No transcript for this episode yet

We transcribe on demand. Request one and we'll notify you when it's ready — usually under 10 minutes.

No similar episodes found.

No similar podcasts found.

Frequently Asked Questions

How long is this episode of Tomayto Tomahto?

This episode is 55 minutes long.

When was this Tomayto Tomahto episode published?

This episode was published on July 3, 2026.

What is this episode about?

Keyon Vafa (postdoc at Harvard) studies the implicit world models that generative models learn (or at least are constrained into representing). How do we make sense of these models and how do they make sense of us? Experimentally, answering this...

Can I download this Tomayto Tomahto episode?

Yes, you can download this episode by clicking the download button on the episode player, or subscribe to the podcast in your preferred podcast app for automatic downloads.
URL copied to clipboard!