Geoffrey Hinton says I'm more optimistic now, not because we'll control AI, but because we might not need to "don't try to dominate superintelligence; design it to care, like a mother wired to protect her child" Control through attachment, not power. we want AI to be like that

It feels the coding agent frontier is now open-weights: GLM 4.5 costs only $3/month and is on par with Sonnet Kimi K2.1 Turbo is 3x speed, 7x cheaper vs Opus 4.1, but as good Kimi K2.1 feels clean. The best model for me. GPT-5 is only good for complicated specs -- too slow.

Introducing K2 Think - a breakthrough in advanced AI reasoning. Developed by MBZUAI’s Institute of Foundation Models and G42, K2 Think delivers frontier reasoning performance at a fraction of the size of today’s largest systems. Smaller. Smarter. Open to the world.

A new dexterous hand is here. DexcelRobotics, a startup founded by a former core member of Tencent Robotics X, has launched its first product, the Apex Hand. The company claims it's the first in the industry capable of operating a cell phone with a single hand. The Apex Hand is

It turned out that model collapse didn't happen. I think there are many reasons to be skeptical of AI lab claims (and point out bad predictions & watch for bubbles) but I also think it is worth reflecting that "AI development is going to stop" arguments have been wrong so far.

DeepSeek V3.2 breakdown 1. Sparse attention via lightning indexer + top_k attention 2. Uses V3.1 Terminus + 1T continued pretraining tokens 3. 5 specialized models (coding, math etc) via RL then distillation for final ckpt 4. GRPO. Reward functions for length penalty, language

🚀 Qwen3-VL-30B-A3B-Instruct & Thinking are here! Smaller size, same powerhouse performance 💪—packed with all the capabilities of Qwen3-VL! 🔧 With just 3B active params, it’s rivaling GPT-5-Mini & Claude4-Sonnet — and often beating them across STEM, VQA, OCR, Video, Agent

Today we're sharing the next phase of Reflection. We're building frontier open intelligence accessible to all. We've assembled an extraordinary AI team, built a frontier LLM training stack, and raised $2 billion. Why Open Intelligence Matters Technological and scientific

Introducing the compact, dense versions of Qwen3-VL — now available in 4B and 8B pairs, each with both Instruct and Thinking variants. ✅ Lower VRAM usage ✅ Full Qwen3-VL capabilities retained ✅ Strong performance across the board Despite their size, they outperform models

Avijit Thawani (Avi) Haha. I am afraid people interpreted my “delete tokenizer” as “use bytes directly without BPE”, the issue is you *still* need bytes encoding arbitrariness even for that! Pixels is the only way. Just like humans. It is written. If GPT-10 uses utf8 at the input I will eat a shoe.

ARC Prize announces all validated scores on ARC-AGI We have not verified MythWorx's 100% claim in their recent fundraise $100M val press release We would be open to verifying their score (assuming it passes the testing policy) for the founder and their investors

Meta laid off 600 people from its Superintelligence Lab today. Many FAIR researchers, including FAIR Research Scientist Director Yuandong Tian, were affected. I think Yann Lecun will leave soon. Maybe I should raise $2B and start a new frontier lab with these folks.

Sakana AI’s CTO says he’s ‘absolutely sick’ of transformers, the tech that powers every major AI model “You should only do the research that wouldn’t happen if you weren’t doing it.” (Brian Cheung) 🧠 Llion Jones venturebeat.com/ai/sakana-ais-…

I disagree with Yann LeCun on this. We have a pretty good idea at Tesla on how we can make general humanoids a reality very quickly. Funny anecdote: Yann was advising me to launch what became the first production vision based deep neural network at Google. His feedback: use convs,

Spiking Neural Network from scratch achieves 8% accuracy. no backpropagation or SGD I created a genetic hyper parameter optimizer and it now, on average, can get 8% accuracy which is ~3% above chance Link to source code with a detailed video and markdown explanations in comment

OpenAI successfully converts from non-profit to un-profitable for-profit.

HRM-Agent: Using the Hierarchical Reasoning Model in Reinforcement Learning Paper: arxiv.org/abs/2510.22832 The Hierarchical Reasoning Model (HRM) has impressive reasoning abilities given its small size, but has only been applied to supervised, static, fully-observable problems.

The trick below to align tokens with different tokenizers is a cute idea -- this allows you to run on-policy distillation with teacher logprobs for sampled tokens even when student and teacher belong to different model families (e.g., Qwen vs Llama). There's more we need to do

FP16 can have a smaller training-inference gap compared to BFloat16, thus fits better for RL. Even the difference between RL algorithms vanishes once FP16 is adopted. Surprising!

🚀Excited to share our new work! 💊Problem: The BF16 precision causes a large training-inference mismatch, leading to unstable RL training. 💡Solution: Just switch to FP16. 🎯That's it. 📰Paper: arxiv.org/pdf/2510.26788 ⭐️Code: github.com/sail-sg/Precis…