Published May 23, 2023
A Psychopathological Approach to Safety in AGI
Assistant Professor Vahid Behzadan delves into the complexities of ensuring AGI safety through psychopathological and behavioral models, addressing the challenges of replicating human cognition and the potential for both cooperation and deception in AGI communication, ultimately aiming for better alignment of AGI objectives with human values.
Topics covered
Popular Clips
Episode Highlights
Related Episodes
AGI Can Be Safe
Answers 383 questions
Uncontrollable AI Risks
Answers 383 questions
Brain Inspired AI
Answers 383 questions
Computable AGI
Answers 383 questionsAI in Games
Answers 383 questions
A Long Way Till AGI
Answers 383 questions
AI Fails on Theory of Mind Tasks
Answers 383 questionsArtificial Intelligence, a Podcast Approach
Answers 383 questionsEvaluating AI Abilities
Answers 383 questions
AI Decision-Making
Answers 383 questionsGame Theory
Answers 383 questions
The Experimental Design of Paranormal Claims
Answers 383 questionsArtificial Intelligence and Auction Design
Answers 383 questions
The PhilPapers Survey
Answers 383 questions
LLMs for Evil
Answers 383 questions