I want to be transparent about how I’ve updated my mainline AI doom scenario in light of safe & useful LLMs. So here’s where I’m at…
00:00 Introduction
07:59 The Dangerous Threshold to Runaway Superintelligence
18:57 Superhuman Goal Optimization = Infinite Time Horizon
21:21 Goal-Completeness by Analogy to Turing-Completeness
26:53 Intellidynamics
29:13 Goal-Optimization Is Convergent
31:15 Early AIs Lose Control of Later AIs
34:46 The Superhuman Threshold Is Real
38:27 Expecting Rapid FOOM
40:20 Rocket Alignment
49:59 Stability of Values Under Self-Modification
53:13 The Way to Heaven Passes Right By Hell
57:32 My Mainline Doom Scenario
01:17:46 What Values Does The Goal Optimizer Have?
Show Notes
My recent episode with Jim Babcock on this same topic of mainline doom scenarios — https://www.youtube.com/watch?v=FaQjEABZ80g
The Rocket Alignment Problem by Eliezer Yudkowsky — https://www.lesswrong.com/posts/Gg9a4y8reWKtLe3Tn/the-rocket-alignment-problem
Come to the Less Online conference on May 30 - Jun 1, 2025: https://less.online
Watch the Lethal Intelligence Guide, the ultimate introduction to AI x-risk! https://www.youtube.com/@lethal-intelligence
PauseAI, the volunteer organization I’m part of: https://pauseai.info
Join the PauseAI Discord — https://discord.gg/2XXWXvErfA — and say hi to me in the #doom-debates-podcast channel!
Doom Debates’ Mission is to raise mainstream awareness of imminent extinction from AGI and build the social infrastructure for high-quality debate.
Support the mission by subscribing to my Substack at https://doomdebates.com and to https://youtube.com/@DoomDebates
Share this post