Simulators - LessWrong

This post explores the concept of simulators in AI, particularly self-supervised models like GPT. Janus argues that GPT and similar models are best u…

x Simulators — LessWrong Simulators Simulator Theory Language Models (LLMs) LLM Personas GPT Outer Alignment Simulation Corrigibility Deconfusion Myopia Oracle AI Tool AI AI Curated 713 Simulators by janus 2nd Sep 2022 AI Alignment Forum Linkpost for generative.ink 49 min read 170 713 Ω 144 Thanks to Chris Scammell, Adam Shimi, Lee Sharkey, Evan Hubinger, Nicholas Dupuis, Leo Gao, Johannes Treutlein, and Jonathan Low for feedback on drafts. This work was carried out while at Conjecture . "Moebius illustration of a simulacrum living in an AI-generated story discovering it is in a simulation" by

saved by

• Alex K. Chen
• Bryan Chiang
• Madison Ueland
• Micah Carroll
• Tazik Sh
• Kevin Wang
• Karan MJ
• Varun Shenoy
• Ratan Kaliani
• Atem Aguer
• Arden Berg
• Lydia Nottingham

related reading

• Janus' Simulatorsastralcodexten.substack.com
• interpreting GPT: the logit lens — LessWronglesswrong.com
• Cyborgism — LessWronglesswrong.com
• Role-playing vs Self-modelling — LessWronglesswrong.com
• Alignment remains a hard, unsolved problem — LessWronglesswrong.com
• You can’t imitation-learn how to continual-learn — LessWronglesswrong.com
• Why Simulator AIs want to be Active Inference AIs — AI Alignment Forumalignmentforum.org
• Model Organisms of Misalignment: The Case for a New Pillar of Alignment Research — AI Alignment Forumalignmentforum.org
• Deep Deceptiveness — LessWronglesswrong.com
• My picture of the present in AI — LessWronglesswrong.com
• Simulators :: — Moiregenerative.ink
• Self-Fulfilling Misalignment Data Might Be Poisoning Our AI Modelsturntrout.com