Robert Long

In this discussion, Robert Long, a philosophy fellow at the Center for AI Safety, examines the contentious topic of AI consciousness. He explains why large language models like GPT likely aren’t sentient entities despite their complex outputs. Long emphasizes the differences between human cognition and AI processing, exploring ethical implications of creating potentially conscious machines. He also addresses the philosophical dilemmas surrounding AI's ability to experience pain and pleasure, urging a cautious approach as we navigate the future of artificial intelligence.

In this episode of 80k After Hours, Luisa Rodriguez and Robert Long have an honest conversation about the challenges of independent research.Links to learn more, highlights and full transcript.They cover:Assigning probabilities when you’re really uncertainStruggles around self-belief and imposter syndromeThe importance of sharing work even when it feels terribleBalancing impact and fun in a jobAnd some mistakes researchers often makeWho this episode is for:People pursuing independent researchPeople who struggle with self-beliefPeople who feel a pull towards pursuing a career they don’t actually wantWho this episode isn’t for:People convinced that their research is perfectAngus Hübelschmidt — the president and sole member of the Rob Wiblin Fan Club who refuses to listen to another Rob speakYou can find their longer conversation on why large language models like GPT (probably) aren't conscious over on the original 80,000 Hours Podcast feed.Get this episode by subscribing to our more experimental podcast on the world’s most pressing problems and how to solve them: type 80k After Hours into your podcasting app.Producer: Keiran HarrisAudio mastering: Ben Cordell and Milo McGuireTranscriptions: Katy Moore

Robert Long, an expert on large language models and consciousness, discusses the ethical implications of AI systems experiencing pleasure and suffering. They explore the possibility of digital minds colonizing the solar system, the therapeutic effects of psychedelics, and whether large language models like GPT are conscious.