The state of AI is a bit of a circus. 1) Every month there's a new model. 2) It's smart, affordable and fast 3) After 1 week max it gets giga-nerfed - token usage, effort, speed. 4) Rinse, repeat. Unfortunately the push is for power, instead of stability and quality.

One more thing: we’re increasing five-hour usage limits on Pro, Max, and Team plans. We’re also providing subscription users a rate limit reset, which you can save and use whenever you choose.

We're adding support for AGENTS.md to Claude Code. Starting today in version 2.1.277, if there is no CLAUDE.md in a folder, Claude will check for and use AGENTS.md. You can toggle this behavior in /config.

We built high-throughput materials labs in Menlo Park to create a loop between experiments and models. The labs generate fresh data, the models learn from it, and then help us decide what to try next. Using only 1,300 H200s, plus months of our experimental data, we mid-trained and RL’d an open-source model to surpass GPT-6 Astra on our analysis benchmark. We call it Neon. This is real footage from our lab. We’re focusing first on hard problems in materials science, including superconductors, magnets, and semiconductor materials. Read our blog posts below.

After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI? I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev • 20-200x faster • 40-400x cheaper (w/ output tokens free) • Frontier composable intelligence optimized for decisions AFAICT the shortest path to AI-based economic revolution

Australian Insurance provider @Zurich is now offering cheaper insurance for Teslas with FSD (Supervised) activated — an Australian first 🇦🇺 Zurich says Tesla data shows FSD-equipped cars have 7x fewer major or minor collisions than regular EVs.

We’re sharing a solution to the Navier-Stokes Millennium Prize Problem, one of the deepest problems at the frontier of mathematics. The proof was produced by a group of agents, using an OpenAI next-generation model significantly more capable than GPT-6 Astra. The problem concerns whether the description of smooth three-dimensional fluid motion modeled by the Navier-Stokes equations can break down. It has remained unresolved for roughly 90 years.

Omarchy Arcade is up on GitHub, with a playable preview if you fancy a go. I want this to become something properly cool, the sort of thing you’re pleased came with your computer, so I’d love some company building it. PRs and issues very much encouraged. Tell me what feels off, suggest a game, make something better. github.com/tcballard/omarchy…

"Hi I'm a Cybercab" "And I'm a Waymo"