There's a new bill, SB-1047 'Safe and Secure Innovation for Frontier Artificial Intelligence Models Act'.

I think it could do a great deal of harm to startups, American innovation, open source, and safety. So I've written a response to the authors: 🧵
answer.ai/posts/2024-04-…

Remembering Daniel Dennett, a beloved figure at SFI known for his work on consciousness and evolutionary biology. He passed away on April 19 at the age of 82. His contributions to philosophical debates were profound & his legacy will continue to inspire.

santafe.edu/news-center/ne…

Very useful thread, if you use Prolific to do human studies.

📣 CALL FOR PAPERS

An ICML Conference workshop on LLMs 🤖and Cognition 💭 (LLMCog)
📍 Vienna July 27

Submit your 4-page papers due May 22
Attend to hear from our invited speakers, the amazing Melanie Mitchell, Subbarao Kambhampati (కంభంపాటి సుబ్బారావు), Chelsea Finn, and Antoine Bosselut !

llm-cognition.github.io

⏰ Time’s ticking! The deadline for the 2024 #ComplexityGlobalSchool has been extended to April 22. 🌎 ✈️ Don’t wait any longer — apply today for the opportunity to dive deep into complexity science. All expenses covered for admitted students.

apply-sfi.smapply.org/prog/2024_comp…

theatlantic.com/technology/arc…

Wrenching news: Dan Dennett has died.

He's been a great friend and incredible inspiration for me throughout my career. I will miss him enormously.

dailynous.com/2024/04/19/dan…

Very good piece on AI risk by Jack Stilgoe

science.org/doi/10.1126/sc…

😲 (but actually not surprising) ⬇️

WTAF?

👀

The New Yorker article, 'Among the AI Doomsayers' by Andrew Marantz is generally pretty good, but recounts the scary-sounding story below.

BUT -- importantly, that's NOT what really happened. For the real (much less scary) story, see my post about it: aiguide.substack.com/p/did-gpt-4-hi…

Quote below is from nytimes.com/2024/04/15/tec…
in NY Times today.

What evidence is there for this claim, besides 'I talked to a chatbot and it seemed humanlike'?

When given an actual Turing Test, chatbots tend to fail (arxiv.org/abs/2310.20216)

Thrilled to share a review on THE LANGUAGE NETWORK AS A NATURAL KIND—a culmination of ~20 yrs of thinking about+studying language from linguistic, psycholinguistic, and cog neuro perspectives. Nature Rev Neurosci rdcu.be/dEylV
With the amazing Anna Ivanova Tamar Regev 🥳
🧵1/n

Highly recommended!

Relying on huge amounts of training data is not sustainable, and it may even be dangerous. This is one of the many reasons why AI should learn from how children develop: they learn so much from so little!