🚀 Mission View
A sharper perspective on this week's issue that matters at the intersection of health and AI.
Trilliant Health’s latest report on the U.S. health economy lays out the same troubling storylines: rising costs, shrinking access, workforce shortages, and uneven value. The authors call the system “unsustainable for patients, payers, employers, and providers alike.” TBH, what’s striking isn’t the diagnosis. It’s what’s missing. The report barely mentions artificial intelligence as a potential solution to some of these problems.
Yet AI may be the biggest wildcard in whether these trends improve or worsen. Used well, AI could extend primary care reach, target waste, and rebalance supply and demand. Used poorly, it could deepen inequities, de-skill clinicians, and inflate administrative complexity. In that sense, AI isn’t a cure-all for the health economy. It’s a stress test for leadership.
The question for mission-driven health organizations isn’t whether to adopt AI, but how to use it to solve inefficiency rather than make things worse. As the tools mature, the challenge will be to prove that faster, smarter systems can also deliver better value.
🛜 Field Signals
A quick hit on this week’s key policy shifts and industry trends.
New research shows AI can accurately screen for diabetic retinopathy (a leading cause of vision loss) before symptoms appear. This could increase screening access in rural areas.
Americans are comfortable with the use of AI in health care settings — as long as it doesn't replace doctors, per the latest Axios-Ipsos American Health Index. By the numbers: 75% of the public is very or somewhat comfortable with AI scheduling appointments. Smaller majorities also are fine with the bots delivering test results (57%) or doing triage and asking basic health status questions (53%).
Axios also reports this week California Gov. Gavin Newsom (D) signed a batch of bills aimed at protecting kids using chatbots and AI tools online.
Amazon Web Services is working with drug companies like Pfizer and AstraZeneca to revamp and accelerate how they make drugs — from research and discovery to manufacturing — using artificial intelligence agents that can reason.
OpenAI assembled an eight-person Expert Council on Well-Being and AI with researchers specializing in psychology, psychiatry, and human-computer interaction to advise on building healthier ChatGPT and Sora experiences.
Senator Edward J. Markey (D-Mass.) introduced the Right to Override Act, legislation that would protect patients and health care workers by requiring an option for human override of artificial intelligence (AI)-based systems that recommend clinical decisions in health care settings.
“The new Dr. Google is in. Here’s how to use it” - the latest Washington Post Opinion by Dr. Leana Wen examining the intersection of health and AI.
WSJ: Microsoft Tries to Catch Up in AI With Healthcare Push, Harvard Deal
U.S. Representative Jay Obernolte (R-CA) launched the AI Coalition to give startups a voice in federal policymaking.
Senators Hawley (R-MO) and Durbin (D-IL) introduced the bipartisan AI LEAD Act to classify AI systems as products, allowing liability claims against developers.
Google and Yale University researchers introduced, a foundation model — based on Google’s open-Gemma — that discovered a previously unknown cancer treatment pathway, proven to work in living cells.
🛠️ Practical Edge
Actionable tips and tools to help leaders strengthen capacity and apply AI in their work.
HBR: Five leadership skills you need in the age of AI.
Google released Veo 3.1 for video generation with powerful new features (at the same price as Veo 3).
Researchers at Stanford and Northeastern found that large language models like ChatGPT are trained to be boring. During training, human raters rewarded safe, predictable answers — so the models learned to suppress their more creative ideas. The fix is surprisingly simple: prompt the model to “generate 5 responses with their corresponding probabilities, sampled from the full distribution.” The result? Far more original writing, dialogue that mirrors real conversation, and greater diversity of ideas — especially on advanced models like GPT-4 and Claude. Hat tip to the folks at The Neuron for flagging this one.
Attrove turns your scattered Slack messages, emails, and calendar events into a single dashboard of prioritized updates, so you can instantly see trending topics and team progress without digging through apps — free tier available, Pro plan with launch discount.
Politico reports how health systems want to adopt artificial intelligence, but they often struggle with selecting the right AI tools and integrating them into their existing digital infrastructure and clinical practice. A new company believes it has a solution to that. Vega Health is a startup that aims to validate, implement and monitor AI algorithms intended for health systems.
The Digital Medicine Society, with support from Google Health and 30 other organizations, has announced a new playbook that aims to help hospitals and health systems craft AI implementation strategies that are grounded in the real needs and capabilities of individual providers.
🌅 On the Horizon
A quick look at the developments and events expected to shape the weeks ahead.
Oct. 21-22: TedAI San Francisco.
Oct. 22-23: PyTorch Conference in San Francisco. Register here.
Nov. 18-20: AI in Healthcare & Pharma Summit 2025 (RE•WORK) in Boston. Register here.
Dec. 8-9: Fortune Brainstorm AI San Francisco. Apply to attend here.
Feb. 11-13: The Complexities of AI in Health Care by American Health Law Association in Las Vegas and Virtual. Register here.
⌚️ Closing Time
A parting thought on what health leaders need to be focused on.
Globant and MIT Technology Review’s new report on agentic AI in pharma makes one thing clear: the next wave of AI isn’t just about R&D efficiency. It’s also about how companies interact with patients. Executives see near-term potential in patient support, adherence, and personalized engagement, where AI could tailor outreach and treatment experiences in real time. Patient advocates take note: the promise is personalization; the risk is substitution. As pharma leans further into AI-driven engagement, the challenge will be making sure “personalized” doesn’t become “automated.” Patients still need relationships, not just AI generated-recommendations.
Till next time,
