AGI - AI Dynamics

Popping this up: a response to a question about what I consider reasoning & planning, why current Auto-Regressive LLMs can't do it, why that would require AI systems with world models, and why we still have a lot of progress to do towards AI systems that can learn and reason.

→ View original post on X — @ylecun

26 November 2023

Scaling Focus Wasted Four Years of AI Progress

By

@garymarcus

–

26 November 2023 20h48

See my 2020 essay The Next Decade in AI, https://
arxiv.org/abs/2002.06177 IMHO we have wasted nearly 4 years since then, blindly chasing fantasies about scaling, and not making much progress on what I suggested there.

→ View original post on X — @garymarcus

26 November 2023

AI System Fails Turing Test Benchmark

By

@garymarcus

–

26 November 2023 20h38

Oops, so much for passing the Turing Test.

→ View original post on X — @garymarcus

26 November 2023

AGI Ruin: A Comprehensive List of Lethalities and Counterarguments

By

@esyudkowsky

–

26 November 2023 20h29

It's not great, but it sure beats the detail levels of any counterclaims: https://
lesswrong.com/posts/uMQ3cqWD
PHhjtiesc/agi-ruin-a-list-of-lethalities
…

→ View original post on X — @esyudkowsky

26 November 2023

When Will AI Systems Reliably Pass the Turing Test?

By

@erikbryn

–

26 November 2023 20h26

What’s your best estimate of when an AI system will reliably pass the Turing test?

→ View original post on X — @erikbryn

26 November 2023

AGI Power: Beyond Greed to Governance Control

By

@alexjc

–

26 November 2023 19h27

OK, challenge accepted! I don't think greed (as pictured) is the main driver for people who stand to benefit from AGI. As a class, they have all the money they want. What they want is the next form of power: governments are decaying, and they need to control what's next…

→ View original post on X — @alexjc

26 November 2023