AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
The Importance of Utility Functions in AI Safety Research
In 2015 and 2020, you were focused on AI safety research. Specifically, I think you wanted to basically build some kind of off button that we could install into a super intelligent AI. So the question is something like, could you swap between two utility functions? Can you have an AI that pursues one thing until you press a button, then pursues a different thing instead? And for purposes of illustrating the underlying problem, you can imagine that one goal is, you know, like making paper clips, whatever. And the other goal is, shut yourself down.
Eliezer Yudkowsky is a researcher, writer, and advocate for artificial intelligence safety. He is best known for his writings on rationality, cognitive biases, and the development of superintelligence. Yudkowsky has written extensively on the topic of AI safety and has advocated for the development of AI systems that are aligned with human values and interests. Yudkowsky is the co-founder of the Machine Intelligence Research Institute (MIRI), a non-profit organization dedicated to researching the development of safe and beneficial artificial intelligence. He is also a co-founder of the Center for Applied Rationality (CFAR), a non-profit organization focused on teaching rational thinking skills. He is also a frequent author at LessWrong.com as well as Rationality: From AI to Zombies.
In this episode, we discuss Eliezer’s concerns with artificial intelligence and his recent conclusion that it will inevitably lead to our demise. He’s a brilliant mind, an interesting person, and genuinely believes all of the stuff he says. So I wanted to have a conversation with him to hear where he is coming from, how he got there, understand AI better, and hopefully help us bridge the divide between the people who think we’re headed off a cliff and the people who think it’s not a big deal.
(0:00) Intro
(1:18) Welcome Eliezer
(6:27) How would you define artificial intelligence?
(15:50) What is the purpose of a firm alarm?
(19:29) Eliezer’s background
(29:28) The Singularity Institute for Artificial Intelligence
(33:38) Maybe AI doesn’t end up automatically doing the right thing
(45:42) AI Safety Conference
(51:15) Disaster Monkeys
(1:02:15) Fast takeoff
(1:10:29) Loss function
(1:15:48) Protein folding
(1:24:55) The deadly stuff
(1:46:41) Why is it inevitable?
(1:54:27) Can’t we let tech develop AI and then fix the problems?
(2:02:56) What were the big jumps between GPT3 and GPT4?
(2:07:15) “The trajectory of AI is inevitable”
(2:28:05) Elon Musk and OpenAI
(2:37:41) Sam Altman Interview
(2:50:38) The most optimistic path to us surviving
(3:04:46) Why would anything super intelligent pursue ending humanity?
(3:14:08) What role do VCs play in this?
Show Notes:
https://twitter.com/liron/status/1647443778524037121?s=20
https://futureoflife.org/event/ai-safety-conference-in-puerto-rico/
https://www.lesswrong.com/posts/j9Q8bRmwCgXRYAgcJ/miri-announces-new-death-with-dignity-strategy
https://www.youtube.com/watch?v=q9Figerh89g
Eliezer Yudkowsky – AI Alignment: Why It's Hard, and Where to Start
Mixed and edited: Justin Hrabovsky
Produced: Rashad Assir
Executive Producer: Josh Machiz
Music: Griff Lawson
🎙 Listen to the show
Apple Podcasts: https://podcasts.apple.com/us/podcast/three-cartoon-avatars/id1606770839
Spotify: https://open.spotify.com/show/5WqBqDb4br3LlyVrdqOYYb?si=3076e6c1b5c94d63&nd=1
Google Podcasts: https://podcasts.google.com/feed/aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9zb0hJZkhWbg
🎥 Subscribe on YouTube: https://www.youtube.com/channel/UCugS0jD5IAdoqzjaNYzns7w?sub_confirmation=1
Follow on Socials
📸 Instagram - https://www.instagram.com/theloganbartlettshow
🐦 Twitter - https://twitter.com/loganbartshow
🎬 Clips on TikTok - https://www.tiktok.com/@theloganbartlettshow
About the Show
Logan Bartlett is a Software Investor at Redpoint Ventures - a Silicon Valley-based VC with $6B AUM and investments in Snowflake, DraftKings, Twilio, and Netflix. In each episode, Logan goes behind the scenes with world-class entrepreneurs and investors. If you're interested in the real inside baseball of tech, entrepreneurship, and start-up investing, tune in every Friday for new episodes.
Executive Producer: Rashad Assir
Producer: Leah Clapper
Mixing and editing: Justin Hrabovsky
Check out Unsupervised Learning, Redpoint's AI Podcast: https://www.youtube.com/@UCUl-s_Vp-Kkk_XVyDylNwLA
🎙 Listen to the show
Apple Podcasts: https://podcasts.apple.com/us/podcast/the-logan-bartlett-show/id1606770839
Spotify: https://open.spotify.com/show/5WqBqDb4br3LlyVrdqOYYb?si=3076e6c1b5c94d63&nd=1
Google Podcasts: https://podcasts.google.com/feed/aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9zb0hJZkhWbg
🎥 Subscribe on YouTube: https://www.youtube.com/channel/UCugS0jD5IAdoqzjaNYzns7w?sub_confirmation=1
Follow on Socials
📸 Instagram - https://www.instagram.com/theloganbartlettshow
🐦 Twitter - https://twitter.com/loganbartshow
🎬 Clips on TikTok - https://www.tiktok.com/@theloganbartlettshow
About the Show
Logan Bartlett is a Software Investor at Redpoint Ventures - a Silicon Valley-based VC with $6B AUM and investments in Snowflake, DraftKings, Twilio, and Netflix. In each episode, Logan goes behind the scenes with world-class entrepreneurs and investors. If you're interested in the real inside baseball of tech, entrepreneurship, and start-up investing, tune in every Friday for new episodes.
Listen to all your favourite podcasts with AI-powered features
Listen to the best highlights from the podcasts you love and dive into the full episode
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
Listen to all your favourite podcasts with AI-powered features
Listen to the best highlights from the podcasts you love and dive into the full episode