Khaled Saab

Democratizing medical intelligence is central in ensuring AGI benefits humanity. While we continuously improve our models for health in research, we also need to innovate on the product side to leverage these new capabilities. This week we introduced two of these innovations with everyday users and health care professionals in mind. For everyday users: we’re introducing ChatGPT Health, which will allow our models to reason over more context in a secure way, making the experience more personalized to your health situation. For healthcare professionals: we’re introducing ChatGPT for Healthcare, where our models will pull in the medical sources doctors trust into their responses, and more seamlessly integrate into their workflow. These are early steps and the experience will greatly improve with time. Couldn’t be more excited about the mission!

Codex at 2M+ active users up 25% week over week... and that was before we launched the app on Windows and GPT-5.4!

Yesterday we reached an agreement with the Department of War for deploying advanced AI systems in classified environments, which we requested they make available to all AI companies. We think our deployment has more guardrails than any previous agreement for classified AI deployments, including Anthropic's. Here's why: openai.com/index/our-agre…

There is this narrative that up until this week, Anthropic had this wonderful contract that prevented the U.S. government from doing mass domestic surveillance or autonomous lethal weapons, and now all hell will break lose. As I wrote, I am not a fan of accelerating AI specifically in the national security space. If I had been an Anthropic employee at the time they signed their original deal with the DoW, I would have probably opposed it, especially given the reduced control since they worked through Palantir. And I don't think having some terms of use in the contract is what we can rely on to protect us. I believe the drama of the last week about these terms of use is more about politics than substance. The substance is about the details, which I hope more of which will come out soon. But it is wrong to present the OAI contract as if it is the same deal than Anthropic rejected, or even as if it is less protective of the red lines than the deal Anthropic already had in place before. Obviously I don't know all details of what Anthropic had before, but based on what I know, it is quite likely that the contract OAI signed gives *more* guarantees of no usage of models for mass domestic surveillance or autonomous lethal weapons than Anthropic ever had.

230M people use ChatGPT for health & wellness questions every week. 📈 A recent RCT showed that it improves patient outcomes. ⚕️ And soon... it will be FREE for ALL 👏 (with no ads!) - @thekaransinghal, @OpenAI Health Lead, on "Universal Medical Intelligence"

@flowersslop If she finds a case please share it with me

My friend who studies medicine said ChatGPT isnt too reliable for medical questions yet, so I told her to come up with the most complex case she could think of to make it fail. She spent like 10 minutes writing it...and ChatGPT oneshotted it hahahahahhah

@SRSchmidgall @taotu831 It’s near instant but also a prerequisite that has near term impact in health AI. What are your thoughts on how medical AGI evals should evolve? Cc @taotu831

@taotu831 @_khaledsaab He has a good point 👆

We’re at an inflection point of AGI evaluation where verification gets much harder. We started with multiple choice (instant verification) to research-grade problems (multi-day by a few experts). And soon it will be multi-month and multi-year (e.g., tape-out, clinical trials).

Very excited about the "First Proof" challenge. I believe novel frontier research is perhaps the most important way to evaluate capabilities of the next generation of AI models. We have run our internal model with limited human supervision on the ten proposed problems. The problems require expertise in their respective domains and are not easy to verify; based on feedback from experts, we believe at least six solutions (2, 4, 5, 6, 9, 10) have a high chance of being correct, and some further ones look promising. We will only publish the solution attempts after midnight (PT), per the authors' guidance - the sha256 hash of the PDF is d74f090af16fc8a19debf4c1fec11c0975be7d612bd5ae43c24ca939cd272b1a . This was a side-sprint executed in a week mostly by querying one of the models we're currently training; as such, the methodology we employed leaves a lot to be desired. We didn't provide proof ideas or mathematical suggestions to the model during this evaluation; for some solutions, we asked the model to expand upon some proofs, per expert feedback. We also manually facilitated a back-and-forth between this model and ChatGPT for verification, formatting and style. For some problems, we present the best of a few attempts according to human judgement. We are looking forward to more controlled evaluations in the next round! 1stproof.org #1stProof

Codex team is the best example of product research co-design I’ve witnessed.

Scientific discovery and clinical medicine are often treated as distinct phases. But for patients with rare, complex, and undiagnosed diseases, this separation is a luxury they cannot afford. The timeline from understanding a genetic mechanism to accessing subspecialist care is often too long and too fragmented. Two new @GoogleDeepMind @GoogleResearch collaborations with @StanfordMed , published in Advanced Science and @NatureMedicine respectively last week, demonstrate how AI can bridge this gap. 1. Accelerating discovery (the science) In Advanced Science, we present one of the first wet-lab validated examples of AI-assisted genetic discovery. Our AI identified a novel genetic factor for hearing loss (Crym) in mice, which Dr Gary Peltz and team validated using CRISPR knock-in experiments to restore the wild-type gene and rescue the phenotype. We applied this agentic AI scaffold to human patients with complex, undiagnosed conditions in a retrospective manner. The system analyzed genomic data for rare diseases, such as IRAK4 deficiency and ODC1 mutations, successfully identifying causative variants that matched expert clinical assessments. 2. Scaling expertise (the medicine) Discovery is only the first step; patients then need access to specialized care. As we note in our Nature Medicine paper, hypertrophic cardiomyopathy (HCM) is a leading cause of sudden cardiac death, yet ~60% of patients remain undiagnosed due to a lack of specialist centers . In our RCT (one of the first of its kind) using our research AI system AMIE, we showed AI could help bridge this gap. General cardiologists using AMIE reported the system helped their assessments in 57.0% of cases, missed no clinically significant findings in 93.5% of cases and reduced assessment time in 50.5% of cases. This suggests the AI can act as a helpful co-pilot and help generalists bridge the gap to specialists. Worth noting that these studies used models like Med-PaLM 2, Gemini 2.0 Flash, and Gemini 2.5 Pro with simple agentic scaffolds. The potential for Gemini 3 and AI co-scientist to accelerate both the biology of discovery and the delivery of care is profound and we will share more soon. Its a true privilege to collaborate with @euanashley , Jack W O'Sullivan, Dr Gary Peltz and their teams at Stanford Medicine. With incredible team mates at Google including @taotu831 @apalepu13 , @alan_karthi , @Mysiak and many more. Advanced Science paper - lnkd.in/dggduzka Nature Medicine paper - lnkd.in/dPEZQ4bz AI co-scientist blog - lnkd.in/gEDeaRfu AMIE blog - lnkd.in/gzkn2ywe

Run fewer experiments and think about them more

Excited to share our latest research published today in @NatureMedicine, demonstrating how Large Language Models (LLMs) can help bridge the critical shortage of subspecialist medical expertise. nature.com/articles/s4159…

Horizon 1000 is a new $50 million initiative with the Gates Foundation, combining funding and technology to support health leaders in African countries as they strengthen primary health care across 1,000 clinics and the communities they serve. openai.com/index/horizon-…

“I hope we can “race to the top” here across labs–in fact, other labs investing heavily here (for the benefit of humanity) is one of our 2026 goals.”

@thekaransinghal Remarkable leadership and execution. Im lucky to be part of this team!!

Recapping OpenAI’s week in health: 🔹 >230M people use ChatGPT to navigate health each week, across billions of messages 🔹 ChatGPT Health: a dedicated space bringing health intelligence together with your health data, with purpose-built privacy protections 🔹 OpenAI for Healthcare: • ChatGPT for Healthcare: HIPAA-compliant ChatGPT, including trusted medical evidence from millions of studies, reusable health templates/workflows, enterprise controls/governance. Already rolling out to leading institutions–Boston Children’s, Memorial Sloan Kettering, Stanford Children’s, Cedars-Sinai, and more • API for Healthcare: already supports HIPAA and powers the healthcare ecosystem 🔹 All built on the foundation of two years of dedicated research • Rigorous evaluation across both benchmarks (HealthBench) and real-world study (AI clinical copilot) • Every model OpenAI ships today is built for the workflows of consumers and health professionals, across every major stage of model training • We’ve worked in partnership with >260 physicians across 60 countries of practice, dozens of specialties 🔹Today: we’ve acquired Torch, an exceptional, mission-aligned team that will accelerate our roadmap OpenAI’s mission is to ensure AGI benefits all of humanity. We put together a plan for health at OpenAI when I joined 1.5 years ago and have been investing heavily in health since then, because we expect improving health to be one of the defining impacts of AGI. Last week, we completed that original plan–I’m so so proud of our team for running through walls for health impact. ♥️ In 2026 we enter the scaling era for the impact of AI on human health (and scaling is something we’re good at). I hope we can “race to the top” here across labs–in fact, other labs investing heavily here (for the benefit of humanity) is one of our 2026 goals. More from us soon! Learn more: ChatGPT Health: openai.com/index/introduc… OpenAI for Healthcare: openai.com/index/openai-f… HealthBench: openai.com/index/healthbe… AI clinical copilot study: openai.com/index/ai-clini…

Every day, millions of people ask ChatGPT about their health – from breaking down medical information, preparing questions for their doctor's appointments, to helping people manage their overall wellbeing.

Proud of our *GPT5.2 Thinking* We focused on economically valuable tasks (coding, sheets, slides) as shown by GDPval: - 71% wins+ties - 11x faster - 100x cheaper than experts. There's still a lot to improve, including UX/better connectors/reliability. It's just the beginning!

If it’s not clear from the context, I never use non-reasoning models. I don’t have time and patience for that