The alignment problem from a deep learning perspective - Ngo et al.
Argument for AI x-risk from competent malign agents - Katja Grace @ AI Impacts
Nate Soares talk @ Google
Leopold I Aschenbrenner
Leopold II Aschenbrenner
Superintelligence as a service
Nathan barnard: “Why I’ve become much less convinced of AI risk”
Nanda overview of threat models Modelling Transformative AI project
Natural Selection Favors AIs over Humans - Dan Hendrycks
Risks from Learned Optimization - Evan Hubinger et. al
My notes on risks from learned optimisation
The basic reasons I expect AGI ruin - Bensinger
Michael Plant Twitter question and replies
Set Sail For Fail? On AI risk - Nintil
Chalmers Twitter question and replies
Eliezer 5-step argument that alignment is hard on Twitter
Survey on AI existential risk scenarios - Sam Clarke
Response to Katja Grace's AI x-risk counterarguments - Johannes Treutlein
[Canonical] Without specific countermeasures, the easiest path to transformative AI likely leads to AI takeover - Ajeya Cotra
[Canonical] Why AI alignment could be hard with modern deep learning - Ajeya Cotra
[Canonical] Is Power-Seeking AI an Existential Risk? Joe Carlsmith
[Canonical] AGI Ruin: A List of Lethalities - Yudkowsky
Collection of arguments to expect (outer and inner) alignment failure? - Sam Clarke
Classifying sources of AI x-risk - Sam Clarke
Quintin Pope against Carlsmith
A Playbook for AI Risk Reduction (focused on misaligned AI) - Holden Karnofsky
Why I am Not An AI Doomer - Sarah Constantin
My Objections to "We’re All Gonna Die with Eliezer Yudkowsky" - Quentin Pope
Against Carlsmith - Thorstad
Sammy Martin, Ben Garfinkel on “Scrutinising AI risk arguments”
[Canonical Counterarguments to the basic AI x-risk case - Katja Grace]
Diminishing Returns in Machine Learning Part 1 - Brian Chau
[Rohit Krishnan]
Scrutinising classic AI risk arguments - Ben Garfinkel
Artificial intelligence is a familiar-looking monster, say Henry Farrell and Cosma Shalizi
To Imagine AI, Imagine No AI - Robin Hanson
Why transformative AI is really, really hard to achieve
[Robin Hanson]
[Tyler Cowen]
https://www.lesswrong.com/tag/object-level-ai-risk-skepticism
Object-Level AI Risk Skepticism - LessWrong
Katja Grace algorithmic progress
Davidson report on explosive economic growth
Transformative AGI by 2043 is <1% likely - Ted Sanders
Will AI cause explosive economic growth? Dietrich Vollrath
Evaluating the historical value misspecification argument
Is the Universal Prior Malign? - Christiano
Worst case guarentees - Christiano
The Solomonoff Prior is Malign - Mark Xu
Without specific countermeasures, the easiest path to transformative AI likely leads to AI takeover - Ajeya Cotra
Takeoff speeds - Paul Christiano Takeoff Speeds and Discontinuities - Sammy Martin and Daniel Eth
Refining the Sharp Left Turn threat model, part 1: claims and mechanisms - Deepmind
DeepMind team - threat models and plans
Threat Model Literature Review - Deepmind
Clarifying AI X-risk - Deepmind [higher level]
Value Learning Sequence - Rohin Shah
Literature Review on Goal-Directedness - Adah Shimi
Deconfusing goal-directedness - Adam Shimi
What is most confusing to you about AI stuff? - Sam Clarke
AI strategy nearcasting - Karnofsky
How might we align transformative AI if it’s developed very soon? - Karnofsky
Important, actionable research questions for the most important century -Karnofsky
Personal thoughts on careers in AI strategy and policy - Carrick Flynn
Reward is not the optimisation target - Alex Turner
Talking about inner and outer alignment - Rohin Shah
The Pointers Problem: Human Values Are A Function Of Humans' Latent Variables - John Wentworth
Evolution provides no evidence for the sharp left turn - Quintin Pope
What Multipolar Failure Looks Like, and Robust Agent-Agnostic Processes (RAAPs) - Andrew Critch
Advanced artificial agents intervene in the provision of reward - [Michael K. Cohen](https://onlinelibrary.wiley.com/authored-by/Cohen/Michael+K.), Marcus Hutter, [Michael A. Osborne](https://onlinelibrary.wiley.com/authored-by/Osborne/Michael+A.)
AGI safety career advice - Richard Ngo
Think carefully before calling RL policies "agents" - Alex Turner
Buck Shlegeris talk the state of alignment
Eli Lifland overview of alignment
How do we become confident in the safety of a machine learning system? - Evan Hubinger
An overview of 11 proposals for building safe advanced AI - Hubinger
Scalable oversight - @anthrupad
Mech interp - @anthrupad
Alignment theory - @anthrupad
Gradient hacking - Richard Ngo
Distillation of “How likely is deceptive alignment?”
How likely is deceptive alignment? - Hubinger
Monitoring for deceptive alignment - Hubinger
Measurement, Optimization, and Take-off Speed
Why alignment is hard, and where to start - Yudkowsky talk
The Rocket Alignment Problem - Yudkowsky
Arbital summary - Yudkowsky
The Sequences - Yudkowsky