Regulate the Flow, Not Just the Source

Percy Liang is a Stanford professor and co-founder of Together AI, driving some of the most critical advances in AI research. 

Percy is also a trained classical pianist, which clearly influences the way he thinks about technology. We explored the evolution of AI from simple token prediction to autonomous agents capable of long-term problem-solving, the problem of interpretability, and the future of AI safety in complex, real-world systems.

[0:00] Intro

[0:46] Discussing OpenAI's O1 Model  

[2:21] The Evolution of AI Agents  

[3:27] Challenges and Benchmarks in AI  

[4:38] Compatibility and Integration Issues  

[6:17] The Future of AI Scaffolding  

[10:05] Academia's Role in AI Research  

[15:17] AI Safety and Holistic Approaches  

[18:32] Regulation and Transparency in AI  

[21:42] Generative Agents and Social Simulations  

[29:14] The State of AI Evaluations  

[32:07] Exploring Evaluation in Language Models  

[35:13] The Challenge of Interpretability  

[39:31] Innovations in Model Architectures  

[43:18] The Future of Inference and Customization  

[46:46] Milestones in AI Research and Reasoning  

[49:43] Robotics and AI: The Road Ahead  

[52:24] AI in Music: A Harmonious Future  

[55:52] AI's Role in Education and Beyond  

[56:30] Quickfire

[59:16] Jacob and Pat Debrief

With your co-hosts: 

@jacobeffron 

- Partner at Redpoint, Former PM Flatiron Health 

@patrickachase 

- Partner at Redpoint, Former ML Engineer LinkedIn 

@ericabrescia 

- Former COO Github, Founder Bitnami (acq’d by VMWare) 

@jordan_segall 

- Partner at Redpoint