What does the Theory of Mind breakthrough discovered in GPT 4 mean for the future of our interactions with language models? How might this complicate our ability to test for AI consciousness? I show the weaknesses of a range of tests of consciousness, and how GPT 4 passes them. I then show how tests like these, and other developments, have led to a difference of opinion at the top of OpenAI on the question of sentience.

I bring numerous academic papers and David Chalmers, an eminent thinker on the hard problem of consciousness, and touch on ARC post yesterday on how they conducted safety evaluations and the urgency of the moment. Featuring Michael Kosinski

Theory of Mind paper: https://arxiv.org/ftp/arxiv/papers/2302/2302.02083.pdf
Faux Pas Results: https://pbs.twimg.com/media/FrcKURnagAIa73i?format=jpg&name=medium
Language Learning Paper: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2884962
Scientific American test: https://www.scientificamerican.com/article/a-test-for-consciousness/
Literature Overview: https://www.researchgate.net/publication/325498266_Reviewing_Tests_for_Machine_Consciousness
Chess Game: https://villekuosmanen.medium.com/i-played-chess-against-chatgpt-4-and-lost-c5798a9049ca
New Scientist Article: https://www.newscientist.com/article/mg20627542-000-picking-our-brains-can-we-make-a-conscious-machine/
P Test: https://www.researchgate.net/publication/228894510_An_Empirical_Framework_for_Objective_Testing_for_P-Consciousness_in_an_Artificial_Agent
Divine Benevolence: https://arxiv.org/pdf/2002.05202.pdf
Slightly Conscious: https://twitter.com/ilyasut/status/1491554478243258368?s=20&t=SRZ7VxYrcXhczjSTwt3W_g
Chalmers Talk: https://www.youtube.com/watch?v=j6cCXg-rjRo
https://en.wikipedia.org/wiki/David_Chalmers
Altman Tweet: https://twitter.com/sama/status/1492645047585570816
Cephalopod Report: https://www.lse.ac.uk/News/News-Assets/PDFs/2021/Sentience-in-Cephalopod-Molluscs-and-Decapod-Crustaceans-Final-Report-November-2021.pdf
Arc Evaluation: https://evals.alignment.org/blog/2023-03-18-update-on-recent-evals/
Michal Kosinski: https://twitter.com/michalkosinski
Bing: Bing.com/new

https://www.patreon.com/AIExplained Non-Hype, Free Newsletter: https://signaltonoise.beehiiv.com/