Paul Christiano
Formerly of OpenAI, stated that a broad understanding of AI capabilities and their evolution reduces risks.
Videos Mentioning Paul Christiano
A.I. Policy and Public Perception - Miles Brundage and Tim Hwang
Y Combinator
Associated with OpenAI, noted for his blog posts on AI safety, including ideas on corrigibility and stable attraction basins for AI systems.
The Origin and Future of RLHF: the secret ingredient for ChatGPT - with Nathan Lambert
Latent Space
Author of the 'Deep Reinforced Learning from Human Preferences' paper (2017), which demonstrated that learning from human preferences could solve basic RL tasks more effectively.
Ilya Sutskever: OpenAI Meta-Learning and Self-Play | MIT Artificial General Intelligence (AGI)
Lex Fridman
Key figure in the OpenAI safety team, involved in work on conveying goals to AI agents.
SmartGPT: Major Benchmark Broken - 89.0% on MMLU + Exam's Many Errors
AI Explained
Formerly of OpenAI, stated that a broad understanding of AI capabilities and their evolution reduces risks.
Eliezer Yudkowsky: Dangers of AI and the End of Human Civilization | Lex Fridman Podcast #368
Lex Fridman
An AI alignment researcher who wrote a response to Yudkowsky's 'AGI Ruin' blog post, suggesting that AI could help expand human knowledge and solve the alignment problem.