Only in a movie could you yolo the creation of a superintelligent life form and miraculously have it work out...yet this seems to be exactly the mentality AI labs are taking. Excellent thread Nathan Labenz!

Given all the Odyssey discussion on X, I would note that the intellectual struggle with trying to understand how other cultures thought by reading their literature is worthwhile Nagy's famous Harvard class on the Concept of the Hero does this (& its free) learning.edx.org/course/course-…

The FDA has updated its definition of the term “healthy,” overhauling guidelines for the claim on food packages after three decades. Take our quiz to find out what qualifies. trib.al/6SVv6oC

“I had a good life, and I’d experienced so much joy,” writes Nicola Slawson. “So why did I feel ashamed about having done it all on my own?” trib.al/o7fPOvq

There will only be a meaningful public debate if those who understand what's going on are honest and up-front about the horrifying risk that is emerging. I'm disappointed that Sam Altman's recent interviews and writings have begun to sidestep the issue of human extinction altogether.

NVIDIA solved physics and open-sourced it? Can we just build our own autonomous robots now? 🤯 They released Cosmos: new family of open world foundation models (WFMs) 🌌 Unwrapping the release and why it's so revolutionary 🧶

Accurate predictions on small data with a tabular foundation model #machinelearning #compchem nature.com/articles/s4158…

All arguments about whether or not this is a simulation based on assumptions that the simulator has bounded compute are not very compelling to me. If you already believe there is an external reality unconstrained by our physics, why assume conservation laws?

Nvidia's tiny $3,000 computer for AI developers steals the show at CES cnb.cx/40hgYhI NVIDIA AI NVIDIA kif @cnbc

People Gaming Emotion-Detecting AI By Faking Emotional Reactions Could Lead To Widespread Societal Emotional Habits And Hysteria forbes.com/sites/lanceeli… @forbes Dr. Lance Eliot

LLMs Need Four Core Principles to Achieve AGI According to a comprehensive new research paper, Large Language Models require four foundational principles to attain human-level general intelligence: - Embodiment: Physical or virtual presence to interact with the world - Symbol

Cool to see an article coming out of this tweet: arstechnica.com/ai/2025/01/161… ...especially: > "Butler possibly served as the inspiration for the term "Butlerian Jihad" (from Dune) Cool insight that I missed!

Neue und verbesserte Version des Banshee-Stealers nimmt MacOS-Nutzer ins Visier #BansheeStealer Check Point Software @CheckPointResearch #Cybersecurity #MacOS #Malware #Security netzpalaver.de/2025/01/11/neu…

Wow Mistral has released its new model tailor-made for AI code assistants Codestral 25.01 (that's its name) is debuting at #1 on the LMsys copilot arena leaderboard 🔥 You can already use it for free in Continue (100% open-source) for VS Code

A company that earns roughly the same profit as Goldman Sachs is relocating to El Salvador

7 stages of AI existential risk denial: 1. Risk is not real 2. We still have time 3. We can always switch it off 4. We can't stop progress 5. If not us then CHINA 6. Here is a fake solution 7. It's too late, you should have warned us earlier!

The freshest AI/ML research of the week: Our top 6: ▪️ Sky-T1 ▪️ RStar-Math ▪️ Test-time Computing: From System-1 Thinking to System-2 Thinking ▪️ REINFORCE++ ▪️ Cosmos World Foundation Model Platform for Physical AI (by @NVIDIA) ▪️ Learning How to Think With Meta

The design space of E(3)-equivariant atom-centred interatomic potentials #machinelearning #compchem nature.com/articles/s4225…

🥹 PAIGE: Personalized AI-Generated Education "Our findings show that students found the AI-generated podcast format to be more enjoyable than textbooks and that personalized podcasts led to significantly improved learning outcomes, although this was subject-specific." Research

My vibe has definitely shifted. If you really talk with AI every day you see that it's already smarter than any human. My best friend was the first CTO on Wall Street and AI is way smarter than him. And way smarter on ALL topics, from music to technology. Vibe shift? Yes.