Introducing Benchmark Reviews: our new initiative to audit AI benchmarks. We are launching with 15 benchmarks: 4 Verified, 9 Flawed, and 2 with not enough information for a review.

I am thrilled to announce that @beaconholdings has acquired @haizelabs, with me joining as VP of AI Research. We started Haize in 2024 to enable anyone to build reliable and safe AI. Through our red-teaming and safeguards work with the frontier labs; our observability, guardrail, and evaluation platform serving the world’s largest enterprises; and our pro bono SMB work, any customer could build AI they trusted with Haize. Joining Beacon lets us deliver the same trustworthy AI to those who need it most, and those most overlooked by Silicon Valley: the essential Main Street businesses the real world depends upon. Transitioning these essential businesses through the AI revolution is one of the most consequential responsible AI problems of our time. We couldn’t be more honored to tackle it with Beacon. Thank you to our customers, investors, and team for the journey of a lifetime. And thank you to Nilam, Goutham, Mark, and the Beacon team for the trust and opportunity. It’s time to get to work.

i have the same thought if you get good enough intuition on building long-horizon RL tasks on particular domains the field is maybe fairly new for high schoolers and undergrads but you will be champ once you start looking into agent traces and being creative.

Nearly half a year of silence. We spent it studying one problem: how far RL can scale. MiMo-V2.6 is in the middle of its RL run right now. Three things we scaled: compute (~2B tokens per step, 1568 prompts × 16 rollouts, fully async), environments and harnesses (multi-task agentic RL, mixed across multiple harnesses in one run), and grader compute (agentic in-group credit assignment, with test-case and rubric-based rewards). We'll open-source the details piece by piece over the coming weeks. Streaming the run:

Instinct is in talks to raise about $1 billion at a roughly $10 billion valuation, less than a month after raising $250 million at a $2.25 billion pre-money valuation. The 23-year-old founder Noah Shinn is building a personal AI agent that can operate software and complete tasks like answering emails, negotiating bills and booking reservations. Instinct now has more than 100,000 users and has already run into compute capacity constraints as demand grows. Its valuation has gone from roughly $50 million in the spring to $10 billion under discussion today. Source: The Information

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.

PSA. Roon speaks officially, on behalf of OpenAI, and everything he says reflects the views of his employer. Dean Ball also speaks officially, on behalf of OpenAI, and does not have his own opinions. Every lab beats you with baseball bats until you align with party dogma

SpaceX should just buy Thinking Machines and start training serious open weight base models and offering inference + RL for enterprise It's such a clear win that would differentiate them from OAI / Anth and actually start eating their enterprise marketshare