Dwarkesh Podcast
Francois Chollet — Why the biggest AI models can't solve simple puzzles
Deeply researched interviews <br/><br/><a href="https://www.dwarkesh.com?utm_medium=podcast">www.dwarkesh.com</a>
Show Notes
Tap timecodes to jump
Here is my conversation with Francois Chollet and Mike Knoop on the $1 million ARC-AGI Prize they're launching today.
I did a bunch of socratic grilling throughout, but Francois’s arguments about why LLMs won’t lead to AGI are very interesting and worth thinking through.
It was really fun discussing/debating the cruxes. Enjoy!
Watch on YouTube. Listen on Apple Podcasts, Spotify, or any other podcast platform. Read the full transcript here.
Timestamps
() – The ARC benchmark
() – Why LLMs struggle with ARC
() – Skill vs intelligence
() - Do we need “AGI” to automate most jobs?
() – Future of AI progress: deep learning + program synthesis
() – How Mike Knoop got nerd-sniped by ARC
() – Million $ ARC Prize
() – Resisting benchmark saturation
() – ARC scores on frontier vs open source models
() – Possible solutions to ARC Prize
Get full access to Dwarkesh Podcast at www.dwarkesh.com/subscribe
This episode is not processed yet. Sign in to queue the transcript and make it useful for search and Q&A.
Sign in to queue transcript (1 credit)
Free to start
Skim the episode first,
Skim the episode first,
then decide if it is worth listening.
Sign up with Google in one click. 10 transcript credits included. No card needed.
Google sign-in · No credit card · 10 free transcript credits