My 5978 Likes

Meet SAM 3, a unified model that enables detection, segmentation, and tracking of objects across images and videos. SAM 3 introduces some of our most highly requested features like text and exemplar prompts to segment all objects of a target category. Learnings from SAM 3 will… https://t.co/qg43OtDyeQ

Introducing Locus: the first AI system to outperform human experts at AI R&D Locus conducts research autonomously over multiple days and achieves superhuman results on RE-Bench given the same resources as humans, as well as SOTA performance on GPU kernel & ML engineering tasks.… https://t.co/W6oVNegAJu

Introducing SAM 3D, the newest addition to the SAM collection, bringing common sense 3D understanding of everyday images. SAM 3D includes two models: 🛋️ SAM 3D Objects for object and scene reconstruction 🧑‍🤝‍🧑 SAM 3D Body for human pose and shape estimation Both models achieve… https://t.co/jLpBX2cY9P

🚨 Stop whatever you're doing. A new study just found that advanced LLMs already think they're more rational than humans and they change their strategy based on who they believe they’re competing against. Yes, you read that right. Researchers built something called the AI… https://t.co/p9KKTuNH9r

You’re looking at neurons growing and connecting in real time 🧠. A 65-hour recording of hippocampal activity in a rat brain. The hippocampus plays a crucial role in memory and learning. In this footage, neurons extend their dendrites and axons, building and reshaping… https://t.co/yeXmmqTKwM

Introducing Gemini 3 Pro for understanding research papers 🚀 Highlight any section of a paper to ask questions and “@” other papers for quick context, comparisons, and benchmark references https://t.co/eLOsYfVSc2

Every PM, designer and founder needs to see this. Replit just shipped an AI designer that outputs professional-grade work in 2 minutes. Design Mode. Powered by Gemini 3 Pro. Launches today. The quality is unreal 👇 https://t.co/Ea810SET1v https://t.co/r8QUlaxEnV

Meet Google Antigravity, your new agentic development platform. An evolution of the IDE, it's built to help you: - Orchestrate agents operating at a higher, task-oriented level - Run parallel tasks with agents across workspaces - Build anything with Gemini 3 Pro. https://t.co/GCwHVJSU1O

This is Gemini 3: our most intelligent model that helps you learn, build and plan anything. It comes with state-of-the-art reasoning capabilities, world-leading multimodal understanding, and enables new agentic coding experiences. 🧵 https://t.co/u9nFSa4h4E

Today, Valar Atomics became the first startup in history to split the atom. Announcing Project Nova, a series of zero power critical tests on Valar Atomics' Nova Core in collaboration with Los Alamos NCERC and NNSS. Nova went critical for the first time this morning at 11:45am. https://t.co/EanSIeaAtS

LOSING DOESN'T EVEN ENTER MY CONSCIOUSNESS https://t.co/5OGfIWdPq2

I just read a paper that completely broke my brain. It describes a system that solved an AI task with over 1,000,000 sequential steps... with ZERO errors. Using AI models that are known to be flaky and make mistakes. How is that even possible? 🤯 We all know LLMs have an… https://t.co/LYqIVQtJzp

Ah! You see, this “May I meet you?”, it is brilliant because it is so stupid. This is the correct stupidity. People say it sounds formal. Good. That is the point. “May I meet you?” is clumsy, yes, but clumsy in the way a real encounter is clumsy. You stumble. You look like an… https://t.co/22vqjeehIl

Exactly 1 year ago, Max and I rebuilt the quantum physics experiment that won the 1997 Nobel Prize in Physics by building a magneto-optical trap which supercools atoms in a room temp, tabletop setup. There are <200 of these across the US and in my time at Open Quantum I'm glad… https://t.co/jkz9ISQqMP

A moment 40 years in the making. Congratulations Tom Cruise on your well-deserved honorary Oscar! https://t.co/f6BQWE1Yrl

"The logic of algorithms tends to repeat what "works," but art opens up what is possible." https://t.co/lFhzgZ8Emq https://t.co/GPRNgkkFE6

One of my favorite lessons I’ve learnt from working with smart people: Action produces information. If you’re unsure of what to do, just do anything, even if it’s the wrong thing. This will give you information about what you should actually be doing. Sounds simple on the…

An idiot in motion will always go further than a genius at rest. https://t.co/LxRP9WaBui

This paper shows, with simple but strong arguments, that there is an unavoidable tradeoff between getting an AI to say fewer wrong things and getting it to handle rare or unusual questions at all. It explains why many popular tests for hallucinations can give a very misleading… https://t.co/RatOxRBHyL

This might be the biggest AI hackathon ever: * >6,300 registrants * Runs for 2 weeks (Nov. 14-30) * Open to anyone, anywhere virtually * $20,000 in cash prizes + $3.5M+ in sponsor credits Hosted by @Anthropic and @Gradio, along with 10 sponsors, join kickoff in 30 minutes 👇 https://t.co/Tz1YTMSKSe https://t.co/Mu3KwZuTTP

https://t.co/cxffVukpJP https://t.co/hvFBiKJalI

I’m seeing AI generated posts and replies everywhere now. ChatGPT has a few telltale signs that give away AI generated content. These will soon be post trained away. It’s not about AI content being everywhere, it's about the fact that we’ll soon lose the ability to tell!

Introducing quickarXiv Papers are often written in convoluted language that is hard to understand. We're fixing that. Swap arxiv → quickarxiv on any paper URL to get an instant blog with figures, insights, and explanations. Now extracted with DeepSeek OCR 🚀 https://t.co/qVXNSjgQBY

Immense planning and technical precision was required for this absolutely preposterous (but real) view: I captured my friend @BlackGryph0n transiting the sun during a skydive. This might be the first photo of it's kind in existence. See a video of this moment in the reply 👇 https://t.co/mkjfavuVsZ

The most powerful scientific instrument of the 21st century isn't the electron microscope or the particle collider. It's the algorithm. Today, a scientist in biology, physics, chemistry etc. is more likely to be debugging a Python script than to be running a wet lab. Going…

SIMA 2 is our most capable AI agent for virtual 3D worlds. 👾🌐 Powered by Gemini, it goes beyond following basic instructions to think, understand, and take actions in interactive environments – meaning you can talk to it through text, voice, or even images. Here’s how 🧵 https://t.co/DuVWGJXW7W

AI writing is everywhere now. It’s showing up in big and little things. Publications, emails, text messages, random speeches I hear at small local events. I cannot unsee it. Seeping into the way people talk and think. Even without AI as a tool, people are going to start…

Introducing Marble by World Labs: a foundation for a spatially intelligent future. Create your world at https://t.co/V267VJu1H9 https://t.co/T00mtETmCA

@laion_ai @Wyndlabs_ai @AmarSVS @0xdrej @bonham_sol @francescodvirga @ChristophSchuh6 @inference_net Blog: https://t.co/OtF9x7kvVc Models: https://t.co/ebaVP7vraN Visualizer: https://t.co/gwjSkWInAi This project spans many disciplines: - bespoke model-training pipelines - high-throughput inference systems - protocols to ensure compute integrity and more. Our goal is… https://t.co/QRRWD3qSHG

Introducing Scribe v2 Realtime – the most accurate real-time Speech to Text model. Built for voice agents, meeting notetakers, and live applications, it transcribes in 150ms across 90+ languages, including English, French, German, Italian, Spanish, Portuguese, Hindi, and… https://t.co/6LcgaHQ9zX