Human generated data has fueled incredible AI progress, but what comes next? 📈 On the latest episode of our podcast, Hannah Fry and David Silver, VP of Reinforcement Learning, talk about how we could move from the era of relying on human data to one where AI could learn... for itself. Watch now → 00:00 Introduction 01:50 Era of experience 03:45 AlphaZero 10:19 Move 37 15:20 Reinforcement learning and human feedback 24:30 AlphaProof 29:50 Math Olympiads 35:00 Experience based methods 42:56 Hannah's reflections 44:00 Fan Hui joinsshow more

Watch → @YouTube Spotify → Apple Podcasts → Or listen wherever you get your podcasts! 🎧

AI is taking the world by storm, going way beyond what we once thought was possible. With investing opportunities galore, it’s time to get in on the action. But how?

@FryRsquared Excited for this one!

@FryRsquared Pumped for this! David Silver is an OG! Not only super accomplished, but a great teacher as well. His RL course was so approachable.

@FryRsquared I really enjoyed the full video—thank you! That said, I found it a bit disappointing that the discussion didn’t directly touch on the recent developments in verification-based reinforcement learning applied to LLMs (like o1, r1, and 2.5 Pro). The focus was mostly on RLHF.

@FryRsquared nice , but whoever did the audio/video sync missed it by 1/10th of a second. be nice with the guy plz, it happens ;)

@FryRsquared It takes a tiny step from MCTS to AGI. Pay attention to 20 Questions - it provides tons of valuable insights for cogsci. Including the core algorithm of cognition. Details in my latest Substack post. I am available for a chat 😉

@FryRsquared Let me translate this video and share it with Chinese people. I’d love to see a Google DeepMind podcast with Professor Hannah Fry.

@FryRsquared AlphaZero ya demostró el potencial del RL. ¿Qué otros problemas complejos podrían resolverse con estos métodos? La transición a IA autodidacta es fascinante, pero ¿cómo garantizar alineación con valores humanos?