Deceptive Alignment

Applied to High-level interpretability: detecting an AI's objectives by Paul Colognese 3d ago

Applied to Understanding strategic deception and deceptive alignment by Marius Hobbhahn 6d ago

Applied to Paper: On measuring situational awareness in LLMs by Owain Evans 1mo ago

Applied to Mesa-Optimization: Explain it like I'm 10 Edition by brook 1mo ago

Applied to Against Almost Every Theory of Impact of Interpretability by Charbel-Raphael Segerie 1mo ago

Applied to Model Organisms of Misalignment: The Case for a New Pillar of Alignment Research by Ethan Perez 2mo ago

Applied to Apollo Research is hiring evals and interpretability engineers & scientists by Marius Hobbhahn 2mo ago

Applied to 3 levels of threat obfuscation by RobertM 2mo ago

Applied to When can we trust model evaluations? by Marius Hobbhahn 2mo ago

Applied to Autonomous Alignment Oversight Framework (AAOF) by Justausername 2mo ago

Applied to Supplementary Alignment Insights Through a Highly Controlled Shutdown Incentive by Justausername 2mo ago

Applied to Disincentivizing deception in mesa optimizers with Model Tampering by martinkunev 3mo ago

Applied to Ten Levels of AI Alignment Difficulty by Samuel Dylan Martin 3mo ago

Applied to A "weak" AGI may attempt an unlikely-to-succeed takeover by RobertM 3mo ago

Applied to Deceptive AI vs. shifting instrumental incentives by Raymond Arnold 3mo ago

Applied to A way to make solving alignment 10.000 times easier. The shorter case for a massive open source simbox project. by AlexFromSafeTransition 3mo ago

Applied to MetaAI: less is less for alignment. by Cleo Nardo 4mo ago

Applied to The Sharp Right Turn: sudden deceptive alignment as a convergent goal by avturchin 4mo ago

Applied to Proposal: labs should precommit to pausing if an AI argues for itself to be improved by NickGabs 4mo ago

Applied to Open Source LLMs Can Now Actively Lie by Josh Levy 4mo ago