All Segments
How to Build AGI? (Ilya Sutskever) | AI Podcast Clips
9 segments available
Full episode with Ilya Sutskever (May 2020): https://www.youtube.com/watch?v=13CZPWmke6A Clips channel (Lex Clips): https://www.youtube.com/lexclips Main channel (Lex Fridman): https://www.youtube.com/lexfridman (more links below) Podcast full episodes playlist: https://www.youtube.com/playlist?list=PLrAXtmErZgOdP_8GztsuKi9nrraNbKKp4 Podcasts clips playlist: https://www.youtube.com/playlist?list=PLrAXtmErZgOeciFP3CBCIEElOJeitOr41 Podcast website: https://lexfridman.com/ai Podcast on Apple Podcasts (iTunes): https://apple.co/2lwqZIr Podcast on Spotify: https://spoti.fi/2nEwCF8 Podcast RSS: https://lexfridman.com/category/ai/feed/ Ilya Sutskever is the co-founder of OpenAI, is one of the most cited computer scientist in history with over 165,000 citations, and to me, is one of the most brilliant and insightful minds ever in the field of deep learning. There are very few people in this world who I would rather talk to and brainstorm with about deep learning, intelligence, and life than Ilya, on and off the mic. Subscribe to this YouTube channel or connect on: - Twitter: https://twitter.com/lexfridman - LinkedIn: https://www.linkedin.com/in/lexfridman - Facebook: https://www.facebook.com/lexfridman - Instagram: https://www.instagram.com/lexfridman - Medium: https://medium.com/@lexfridman - Support on Patreon: https://www.patreon.com/lexfridman
Segments Timeline
The Role of Self-Play in AGI Development
Ilya Sutskever discusses the importance of self-play in building Artificial General Intelligence (AGI). He explains how self-play mechanisms allow systems to learn and improve through competition, leading to surprising and creative solutions to problems. This segment highlights the potential of self-play as a crucial component in developing human-level intelligence.
"what do you think it takes to let's talk about AGI a little bit what do you think it takes to build a system of human level intelligence we talked about reasoning we talked about long term memory but ..."
Simulation vs. Real-World Learning
In this segment, Sutskever addresses the debate between learning in simulated environments versus real-world scenarios for AGI. He emphasizes that while simulation is a valuable tool with its strengths and weaknesses, successful transfer of learning from simulation to the real world is possible, citing examples like OpenAI's robotic hand trained entirely in simulation.
"ways for example like we I mean pretty much every self plays system both our daughter bought I don't know if opening I had a release about multi-agent where you had the two little agents were playing ..."
The Necessity of a Body for AGI
Sutskever explores whether an AGI system needs a physical body to achieve human-like intelligence. He argues that while having a body is beneficial for learning certain experiences, it is not strictly necessary. He references examples of individuals who have compensated for physical limitations, suggesting that AGI could similarly adapt without a body.
"I think simulation is a tool and it helps you has certain its strengths and certain weaknesses and we should use it yeah but okay I understand that that's uh that's true but one of the criticisms of s..."
Consciousness and AGI
This segment delves into the complex topic of consciousness in AGI systems. Sutskever discusses the challenges of defining consciousness and whether it could emerge from advanced neural networks. He posits that if artificial neural networks are sufficiently similar to the human brain, they might also exhibit consciousness, raising profound questions about the nature of intelligence.
"and the better the transfer capabilities are the more useful simulation will become because today and you could take you could experience something in simulation and then learn a moral of the story wh..."
Defining Intelligence in AI
Sutskever reflects on what constitutes intelligence in AI, referencing the Turing Test and the importance of systems performing tasks without human-like errors. He expresses his desire to see AI systems that can handle pedestrian tasks flawlessly, highlighting the current limitations of deep learning models and the skepticism surrounding their intelligence.
"existence proof pretty heavily okay but this is that that's that's the best answer I can give no I I know I know I know there's still an open question if there's not some magic in the brain that were ..."
Public Perception of AI Progress
In this segment, Sutskever discusses the public's perception of AI advancements and the tendency to focus on failures rather than successes. He emphasizes the need for a balanced view of AI's capabilities and the importance of recognizing the impressive achievements of current AI systems, despite their occasional mistakes.
"is progress yeah that's a really nice way to put it but I also just don't like that human instinct to criticize a model is not intelligent that's the same instinct as we do when we criticize any group..."
The Future of AGI and Human Control
Sutskever shares his vision for a future where AGI systems operate under human control, likening it to a board of directors overseeing a CEO. He discusses the importance of relinquishing power and ensuring that AGI systems are designed to assist and flourish alongside humanity, fostering a collaborative relationship.
"this connects to the earlier point we discussed that it's just confusing to judge progress in AI yeah and you know you have a new robot demonstrating something how impressed should you be and I think ..."
The Ideal Relationship with AGI
In this segment, Sutskever elaborates on the ideal relationship between humans and AGI, envisioning a system that is designed to help humanity thrive. He discusses the potential for AGI to be programmed with a desire to assist humans, drawing parallels to parental care and support.
"what do you think after spending that evening having a discussion with the AGI system what do you think you would do well the ideal world would like to imagine is one where humanity I like the board t..."
Aligning AGI Values with Human Values
Sutskever addresses the critical issue of aligning AGI values with human values. He discusses the importance of developing a reward function for AGI that reflects human judgments and values, suggesting that this alignment is essential for creating beneficial AI systems that operate in harmony with human interests.
"be to help humans flourish but let me take a step back to that moment where you create the AGI system I think this is a really crucial moment and between that moment and the the Democratic board membe..."