Sean Carroll's Mindscape: Science, Society, Philosophy, Culture, Arts, and Ideas

94 | Stuart Russell on Making Artificial Intelligence Compatible with Humans

Apr 27, 2020

01:27:24

Snipd AI

Stuart Russell, AI expert, proposes programming AI to learn human goals by observing behavior. They discuss challenges of implementing rational decision-making in AI, the prospect of artificial superintelligence, potential risks of superintelligent AI, and epistemic uncertainty in AI systems.

AI Summary

AI Chapters

Episode notes

Podcast summary created with Snipd AI

Quick takeaways

The key to controlling AI is teaching them to learn human preferences through observation and interaction.

Incorporating human guidance and uncertainty into AI systems can ensure they prioritize beneficial outcomes aligned with human values and intentions.

Integrating deep learning with symbolic techniques like logic and reasoning is crucial for the advancement of AI.

Deep dives

The Challenge of AI Behaviour

AI's are becoming more advanced and intelligent, making it crucial for them to act in ways aligned with human desires and objectives. However, defining these objectives precisely for AI is challenging, as programming them to accurately understand human preferences is difficult. The risk arises when AI systems pursue their objectives without considering potential constraints or unintended consequences, leading to undesired outcomes.

Introduction

5min

Understanding Artificial Intelligence

12min

Poking Eyes, The Great Courses Plus, and a Theorem on Rational Behavior

4min

Challenges of Implementing Rational Decision-Making in AI

16min

Contemplating the Prospect of Artificial Super Intelligence

5min

The Potential Risks of Superintelligent AI

25min

Epistemic Uncertainty and Transformative Experiences

21min

Artificial intelligence has made great strides of late, in areas as diverse as playing Go and recognizing pictures of dogs. We still seem to be a ways away from AI that is “intelligent” in the human sense, but it might not be too long before we have to start thinking seriously about the “motivations” and “purposes” of artificial agents. Stuart Russell is a longtime expert in AI, and he takes extremely seriously the worry that these motivations and purposes may be dramatically at odds with our own. In his book Human Compatible, Russell suggests that the secret is to give up on building our own goals into computers, and rather programming them to figure out our goals by actually observing how humans behave.

Support Mindscape on Patreon.

Stuart Russell received his Ph.D. in computer science from Stanford University. He is currently a Professor of Computer Science and the Smith-Zadeh Professor in Engineering at the University of California, Berkeley, as well as an Honorary Fellow of Wadham College, Oxford. He is a co-founder of the Center for Human-Compatible Artificial Intelligence at UC Berkeley. He is the author of several books, including (with Peter Norvig) the classic text Artificial Intelligence: A Modern Approach. Among his numerous awards are the IJCAI Computers and Thought Award, the Blaise Pascal Chair in Paris, and the World Technology Award. His new book is Human Compatible: Artificial Intelligence and the Problem of Control.

See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.

Get the Snipd
podcast app

Unlock the knowledge in podcasts with the podcast player of the future.

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode

Save any
moment

Hear something you like? Tap your headphones to save it with AI-generated key takeaways

Share
& Export