for those feeling dizzied by the rapid progress and tumult in modern AI, take solace in some of my favorite quotes from everyday folks living through the industrial revolution in 18th century London:

Please steal my AI research ideas. This is a list of research questions and concrete experiments I would love to see done, but don't have bandwidth to get to. If you are looking to break into AI research (e.g. as an undergraduate, or a software engineer in industry), these are

Since launching our AI for Science program, we’ve been working with scientists to understand how AI is accelerating progress. We spoke with 3 labs where Claude is reshaping research—and starting to point towards novel scientific insights and discoveries. anthropic.com/news/accelerat…

Tonight at the house: An intimate AI x Physics Dinner with Prof. Steve Hsu from Manifold steve hsu, Alex Morisse, Alex Morisse, PhD, Jasper Jasper, and @jvnixon. We started off with Alex casually explaining the science behind his stealth deeptech start-up on physics textbooks. A

Talented physics inventors and researchers in the house! Thanks to everyone who is coming to discover something novel in physics using AI. Aditya P. Advani Armin Hamrah Bhavye Khetan bidhan Emanuele Fenocchi @inservixeofx Gautam Goswami Adam (هومن) Ilia Zintchenko @guanniqu

🚀 Introducing Qwen3-Max-Thinking, our most capable reasoning model yet. Trained with massive scale and advanced RL, it delivers strong performance across reasoning, knowledge, tool use, and agent capabilities. ✨ Key innovations: ✅ Adaptive tool-use: intelligently leverages

Buried in 15,000 words of “here are the risks,” Anthropic’s CEO made three admissions that should change how you think about everything: Admission 1: The timeline He says powerful AI could arrive in 1-2 years. He’s watching internal model progress and says he can “feel the pace

Kimi 昨天发布 Kimi K 2.5，它的一些基本信息： 1. MoE 模型，拥有 1T 总参数和 32B 激活参数。 2. 采用原生 INT4 精度发布，而不是 FP8/BF16。 3. 原生支持图像输入和视频输入。 4. 是个混合推理模型，支持推理和非推理模式。 Artificial Analysis 对 Kimi K2.5 的评测结果也出来了，超越GLM

Everyone’s missing the real story here. Anthropic just made the same product move OpenAI made 112 days ago. Same protocol (MCP). Same partners (Figma, Canva, Slack). Same “interactive tools inside chat” framing. The surface narrative: Claude is catching up. The actual

Big respect for Dario, but the essay skips a crucial reality: open-weight models are beating SOTA, and China is leading that wave.

I'm Boris and I created Claude Code. I wanted to quickly share a few tips for using Claude Code, sourced directly from the Claude Code team. The way the team uses Claude is different than how I use it. Remember: there is no one right way to use Claude Code -- everyones' setup is

Jensen is doing damage control in real time and nobody’s pricing it in. The tweet presents the Nvidia bull thesis: CUDA lock-in, NVLink scaling, $20B R&D moat. All true in 2023. Here’s what happened in the last 90 days: Anthropic announced Claude Opus 4.5 was trained primarily

The pace at which industry is evolving 🤯

x.com/i/article/2019…

We've been working on very long-running coding agents. In a recent week-long run, our system peaked at over 1,000 commits per hour across hundreds of agents. We're sharing our findings and an early research preview inside Cursor.

opus 4.6 with new “swarm” mode vs. opus 4.6 without it. 2.5x faster + done better. swarms work! and multi-agent tmux view is *genius*. insane claude code update.

This is a dopamine loop, and it’s one of the most powerful ones humans have ever encountered. Every time you prompt an AI and get a useful result back in seconds, your brain gets a hit. Variable-ratio reinforcement, same mechanism as slot machines, except the reward is real:

Today we share a technical report demonstrating how our drug design engine achieves a step-change in accuracy for predicting biomolecular structures, more than doubling the performance of AlphaFold 3 on key benchmarks and unlocking rational drug design even for examples it has

x.com/i/article/2021…