AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
6min chapter
#364 - Stuart Russell - The Terrifying Problem Of AI Control
Modern Wisdom
How to Get Artificial Intelligence Wrong?
The current approach to a i, which has been there really since the beginningad in the book human compatible, i call it the standard model. The problem is that we don't know how to specify the objective completely and correctly. And so for artificially defined problems, like chess, it kind of fooling us into thinking that is an easy problem to solve. But if you prioritize safety above everything else, then you're never going to leave the garage because just going out on to the road incurs some risk. You can see things go wrong really fast, such as curing cancer or fixing carbon dioxide levels. We have to tweak all the objectives to getting to behave better - but even for that
00:00
Transcript
Episode notes
Get the Snipd
podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered
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
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
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