Proud of my teammates for their rigorous work on MentalHealthBench. They collaborated with over 80 licensed psychologists and psychiatrists to cover diverse scenarios with high-quality rubrics. A great one to saturate!

I'm excited to announce our work on MentalHealthBench, a new open benchmark developed with mental health experts to evaluate how Al models respond in realistic mental health conversations. 🧵

Today we announced the Claude-led discovery of a molecular machine that we suspect could represent a new gene editing mechanism. Its precise function, biotechnological utility (if any), or level of significance is not yet clear, but at minimum it is work I would have been proud to do as a PhD student. The work was done mostly, though not entirely, by Claude: our life sciences team suggested a broad area of research, Claude read through the literature and a bunch of genome data and discovered something interesting, then Claude proposed experiments to verify the discovery and our team carried them out. It’s easy to dismiss this as a one-off or curiosity, but we’ve repeatedly seen a pattern where AI performance in new intellectual domains goes from weak to superhuman in a matter of a few years. In 2023 models struggled to do math at the level of an average high-school student. In 2024 they started to do well on math competitions for the best high-schoolers in the country, in 2025 they started to solve minor open problems, in early 2026 more significant open problems, and in late 2026 they are beginning to solve the top few open problems in all of mathematics. We believe AI for biology is on a similar exponential trend. The main difference between biology and mathematics, of course, is that math can be done purely theoretically, while biology requires experimentation. Some have used this to draw the conclusion that AI’s utility in biology will be limited. We think this is wrong. As we’ve demonstrated today, humans can collaborate with AI to perform the experiments, validate key results in a few weeks and, if necessary, work with the AI to iterate on what they find. Eventually it may even be possible for Claude itself to safely perform the experiments by autonomously controlling lab equipment, with appropriate safeguards in place, but we aren’t doing that today (our lab is also a BSL1/BSL2 facility that doesn't handle materials dangerous to humans). More broadly, biomedical advancement has many stages — from fundamental biology discoveries, to translational research, to drug discovery, clinical trials, and finally the actual delivery of medicines and health care to patients. We are also interested in these later stages, but even simply accelerating the first stage of fundamental biological discoveries has the potential to speed up and broaden the entire pipeline. Improving our understanding of biology and sharpening biologists’ tools can drive forward all of the later stages, for example by identifying new drug targets, finding new therapeutic modalities, allowing for more precise measurement, and speeding up the experimental loop which itself further accelerates our understanding of biology. This will not in itself speed up clinical trial times, but if it succeeds it could greatly increase the number of promising candidates that go into the pipeline — an increase in throughput even though latency remains. In Machines of Loving Grace, I wrote about AI’s potential to “cure most diseases in 5-10 years” — a goal that sounds impossible, but one I believe is just barely possible if AI is applied to every stage of the pipeline. The first step is showing that AI can first help with, and then drive, biological discoveries. Claude’s discovery is the latest in a line of related prior work that goes back decades, beginning with systems like CRISPR, and continuing with discoveries like the bridge recombinase and VIPR in the past few years. Recently, there has been heightened interest in systems based on reverse transcriptase (RT) enzymes, the enzyme underlying the system Claude identified. And most recently, a Stanford team working independently described a novel RT system with an associated non-coding array that is in some ways similar to the one Claude found, though they are distinct systems that evolved independently from each other. I believe that we’re at the very beginning of finding such systems and developing them into powerful tools for biotechnology. I’m proud of the resources Anthropic has invested in accelerating the public benefits of AI through the life sciences, and we’re aiming both to grow our life sciences team and to work with other scientists to extend this approach to a broad range of problems. If you have a proposal for a research collaboration or are interested in joining our life sciences team, please reach out.

We’re demonstrating how frontier models have continued to improve in realistic mental health conversations with MentalHealthBench. This new open benchmark was built with input from more than 80 mental health clinicians. We’re releasing it openly so other researchers can examine the methods, run their own evaluations, and build on the work.

A common trend is that once we measure something well, the models will saturate the performance on that metric within a year. I really hope this happens for model performance in mental health situations! Also congrats Luna, model powering billions of users and topping the chart!

Millions of people turn to AI to navigate difficult relationships, work through stress, or support someone they care about. We’re introducing MentalHealthBench, an open benchmark developed with mental health experts to measure how well AI responds in these moments. 🧵

This release builds on our recent work to make trustworthy health intelligence widely accessible, including the following in the last few months: Free GPT-6 Astra Pro for U.S. clinicians:

GPT-6 Luna and Sol are a step towards health intelligence too cheap to meter. ♥️ These models push the frontier of performance per dollar on HealthBench Professional. GPT-6 Luna scores higher than GPT-5.6 Sol–our best model a few weeks ago–at their highest reasoning efforts, despite GPT-6 Luna being ~34x cheaper: about a third of a cent per response. HealthBench Professional consists of real clinician tasks, including care consultations, medical documentation, and research. This evaluation is designed to be challenging and focuses on especially complex and rare situations. I’m excited for what abundant health intelligence can unlock for patients and clinicians.

Helpful (and dare I say fun?) new feature in ChatGPT Health: tap on anything in your synced data and get a helpful summary of everything related to that metric. No prompt required! Makes use of your chat-wide memory, so can be incredibly personalized and specific to you.

In light of the progress in mathematics, we at Edison Scientific and FutureHouse have assembled a set of Millennium Problems for Biology. They are chosen to be very hard to solve but very easy to validate in a simple laboratory environment. Any of these, if solved, would mark a major advance in biotechnology, and most of them would contribute materially towards curing disease. These are, in some sense, the “last reasonable eval” for AI in biology. This was work primarily by @MichaelaThinks and myself, with contributions from many others. Short descriptions below. The full descriptions of the problems with acceptance criteria are at the Bio Millennium Problems website, linked in the next post. Share more if you have ideas. If they meet our criteria, we’ll add them to our list (with attribution and permission).

Today we’re excited to launch Public Data for Health, with more than $125M in initial grants to nonprofits and universities creating and opening up high-quality scientific datasets. AI is getting dramatically better at analyzing biology, but accelerating medical discovery depends on researchers having high-quality data to work with. Our goal is to make more valuable scientific data broadly available so researchers can build on each other’s work and accelerate discoveries that help prevent and treat disease. Learn more:

My name is Chris Painter, and I'm the President of METR (Model Evaluation and Threat Research). I know we've made a lot of new friends on the internet the last couple of days, so I thought I'd take this chance to re-up what we do and why. Our work is aimed at making sure that if AI really were autonomous, difficult to steer, and close to "going rogue," the public would find out. If evidence exists inside of an AI company that it’s close to losing control of AI, we want to make sure that information gets shared with the rest of the world, including governments and the public outside the company’s walls. This is what we've been focused on since 2022, and over the years we've worked with OpenAI, Anthropic, Google DeepMind, Meta, Amazon, and others on piloting third-party assessments and investigations of this type. We don’t have some private room where we rubber stamp things as “safe” or not. We have had a track record of publishing results on AI that don't cleanly map onto the "doomer" or "accelerationist" labels, and we put in effort to hire people with competing views on AI. We’ve been cited for having found some of the strongest evidence that AI capabilities are improving rapidly (our work measuring AI “time horizons”) while also presenting some of the strongest evidence that, at various points, AI’s capability may be overstated (some might remember our study showing that early 2025 software engineers were actually being slowed when they thought they were being sped up). METR is funded by donations. We don't accept money from frontier AI companies. They haven't paid us for our work, and we don't accept donations from them or their employees. As we’ve shared previously, multiple frontier AI companies currently provide us with free access to their models in order to perform our evaluations, research, and engineering. Our funding intentionally comes from a wide range of donors, which we’ve shared on our website. Today, when an AI company works with any third-party evaluator or external testing organization (of which there are and should be many), it's entirely voluntary. This often involves NDAs and redactions. To counterbalance this, we have a principle that when we enter into a contract with a company, we try to retain the right to tell the public the terms of the contract we signed, and characterize the nature of redactions that the company chose to make. For example, the report from our independent investigation of the OpenAI-HuggingFace incident included that information. Public disclosure is also a big part of our COI policy (linked on our website). That’s not to say our reports are adequate as oversight. We’re just one organization (among many doing great work), working in a voluntary setup, trying to get good evidence to the public and the world about AI, letting the facts fall where they may.

We're sharing our new framework for tracking, investigating, and disclosing instances of model misalignment at OpenAI. The framework sets criteria and timelines for public disclosure, including when we haven’t yet fully explained or mitigated the behavior. More complex cases may require longer investigation or coordination with third parties. We’ll prioritize examples that reveal new misalignment mechanisms, meaningful changes in known behavior, or findings that challenge assumptions about safety or mitigation. Alongside the framework, we’re publishing six reports on instances of misaligned behavior we’ve observed during the training or evaluation of our models in the last six months. This is a starting point. We’ll refine the process through experience and public feedback, and share more reports on an ongoing basis.

GPT-6 Astra is a big step forward in democratizing health intelligence. My teammates and I have been hard at work pushing health throughout the training stack, and we’re excited to ship improvements at cheaper prices with every model generation.