Francois Chollet — Why the biggest AI models can't solve simple puzzles
Dwarkesh Podcast
00:00
Skepticism and the Path to AGI: Evaluating LLM Performance
This chapter explores doubts about large language models reaching an 80% benchmark on a specific test within a year. It emphasizes the significance of genuine understanding over memorization and discusses the adaptation to novel tasks as a key element in the pursuit of artificial general intelligence.
Transcript
Play full episode
