Sponsored by: Champ AI
Eligibility misses don’t cost minutes. They cost months
Plan changed mid-month: claim denied
Deductible already met: refund plus card feeds
Drug already purchased: payment denied
Champ does what your eligibility tool can't. AI agents work the portal, wait on hold with the payer, and get the full answer before the visit.
Choose a portal. Get your first 50 eligibility checks free. Learn more here!
Good morning, readers —
We're excited to announce the “Health Frontier NYC Hackathon”, co-hosted with General Catalyst and Percepta! On Dec 4–5, we're bringing together ~100 of the best builders in AI and healthcare, alongside an incredible lineup of industry partners, to build the future of health and bio. Applications close Nov 24. Apply here!
Now, let's get to it.
What the midterms mean for health AI
The rise of longevity and lifestyle medicine
Healthcare is the largest vertical AI market
48 tools, partnerships, research, policy updates & deals, plus AI jobs and more
Read time: 7 minutes
Our Picks ✨
Highlights if you’ve only got 2 minutes…
1/
What the midterms mean for healthcare AI
Democrats won control of both the House and Senate in the midterms, with healthcare affordability playing a major role. In battleground Senate races alone, Democrats spent $65M on healthcare-focused ads.
Their early agenda includes restoring ACA subsidies, rolling back Medicaid cuts, increasing oversight of insurers and federal health agencies, and ending an AI-powered prior authorization program in Traditional Medicare. With President Trump still in the White House, big reforms will be difficult, so targeted bipartisan action around AI, cybersecurity, price transparency, drug development, and prior auth may be more realistic.
There is an interesting tension here for healthcare AI. Policymakers are looking for ways to make care more affordable while political resistance to AI itself is growing. Voters backed limits on data centers in 40 of 41 local ballot measures, and 76% said they would oppose a new data center in their community. Bernie Sanders and Greg Casar have also proposed pausing advanced AI development until federal safety rules are in place and permanently banning artificial superintelligence.
At the same time, AI is one of the most promising tools for dealing with the cost issues policymakers are trying to solve. Automating administrative work, expanding clinicians’ capacity, and delivering more care without adding human labor at the same rate are a big part of the promise.
So health AI heads into 2027 with more adoption, more scrutiny, and a pretty important policy question: can Washington put guardrails around harmful uses of AI without slowing down the ones that could actually make healthcare cheaper and more accessible?
That balance matters beyond healthcare too. The US has a strategic interest in maintaining its lead in AI, and healthcare is one of the places where the upside could be most tangible. We should be able to put real guardrails around harmful uses without treating every application of AI the same. The goal should be to keep building health AI that expands access, lowers costs, and improves care.
2/
The rise of longevity and lifestyle medicine
Two new reports from a16z and McKinsey point to the same trend: health optimization is moving from the fringe into the mainstream.
a16z’s report calls it the “healthmaxxing economy.” In its survey, 54% of consumers use a wearable, 70% filter their water, and nearly three in four wearable users said they would wear a second device if it measured something new. The bigger opportunity is what they call a “healthcare home” that combines records, labs, wearables, care, and AI, then helps people actually do something with all that information.
McKinsey’s report puts the global wellness market at $2.7T, including roughly $700B in the US. More than a third of Americans now say wellness is core to their identity, and more than 40% say they are willing to “do whatever it takes” to live as long as possible.
Traditional healthcare has largely ignored a lot of this. We’ve built a multitrillion-dollar industry around treating disease downstream, while many of the biggest drivers of health sit outside the traditional healthcare system. Diet, sleep, movement, environment, stress, and social connection all shape health long before the doctor visit. Glad that shift is finally gaining traction.
Longevity, diagnostics, wearables, nutrition, fitness, and lifestyle medicine are becoming part of how people manage health every day. a16z also points to emerging areas like regulated peptides, non-invasive neurotechnology, preventive cardiovascular care, and tools for managing environmental exposures.
AI could make all of this much more useful by connecting the data, understanding someone’s health over time, and helping turn information into action. The next big consumer health company might be a continuous operating system for your health, connecting all of these pieces. (tweet)
3/
Healthcare is the largest vertical AI market
Menlo Ventures’ latest enterprise AI report puts some real numbers behind how quickly AI spending is scaling. Enterprise deployment spend reached $177B in 2026, up 5.4x from last year, with agents driving a growing share of that usage.
Coding remains the runaway leader at $34B, followed by enterprise chatbots at $25B. Healthcare is now third at $3.7B, up 2.5x year over year.
Within industry-specific AI, though, healthcare is way out front. It accounts for 37% of all vertical AI spending and is now larger on its own than the entire vertical AI market was last year. Clinical products drove much of the growth, including point-of-care tools like OpenEvidence and ambient documentation companies like Abridge. Menlo also flags biology and drug discovery as the next major frontier.
It’s crazy because all of this still feels early. Healthcare represents only about 5% of total AI application spending despite being one of the largest industries in the economy.
Menlo’s broader argument is that agents work best when they can access the systems where work happens, understand the relevant context, and verify whether they completed the job correctly. Healthcare still has plenty of work to do on all three. If those pieces improve, $3.7B is more of a starting point than an endpoint. (link)(twitter)
Healthcare AI Rundown 🔍
Latest on tools, partnerships, research and policy…
Tools & Product Updates 🔧
OpenEvidence launches free medical education platform: Learn combines board preparation, case-based AI tutoring, flashcards, and CME tools to support medical trainees and clinicians throughout their careers. (link)
Mayo brings AI clinical guidance into the EHR: Its clinical decision support platform combines specialty-specific chat across 35+ specialties with medical literature, daily updates, and 22 clinical calculators for Mayo clinicians and participating organizations. (link)
Hippocratic AI upgrades patient conversations with Polaris 6 Ultra: The company says its latest model better interprets emotional cues, reasons while speaking, and adapts its voice in real time, with a 1.3x speed improvement over Polaris 5. (link)
HCA targets 10,000 physicians using ambient AI by year-end: The health system says the technology saves approximately 5 minutes per note, potentially returning 1–2 hours per shift as deployment expands. (link)
GE HealthCare adds AI lung and heart assessment tools: Its FDA-cleared Lung Guidance and AutoAS capabilities expand Vscan Air’s handheld ultrasound functionality, supporting lung imaging and aortic stenosis assessment at the point of care. (link)
Glass 6.0 makes clinical AI and ambient scribing free: NPI-verified U.S. clinicians receive unlimited queries and ambient scribing. Its new Agent Customization feature lets clinicians tailor responses and documentation to their preferences, templates, and reference materials. (link)
AI-native laboratory automates complex nucleic acid experiments: AutoDNA coordinated 20+ instrument types across more than 11,000 experimental steps, demonstrating autonomous laboratory workflows with human involvement limited to maintenance. (link)
OpenAI launches benchmark for everyday wet-lab reasoning: BioBasicsBench tests AI agents on interpreting raw experimental data across 100 routine laboratory tasks, extending scientific evaluation beyond textbook questions. (link)
Doximity expands clinical AI to 190+ health systems: Its customer base grew from 165 organizations last quarter and includes 10 of the nation’s top 20 hospitals, as health systems integrate its clinical AI tools into existing workflows. (link)
Neurotronics receives FDA clearance for home sleep analysis: Its prescription-only Sleep Analysis Module uses data from home sleep apnea tests to estimate sleep duration, adding AI-supported analysis to testing performed outside a sleep laboratory. (link)
HOPPR releases AI model for musculoskeletal X-ray reporting: The beta model generates draft findings for whole-body musculoskeletal radiographs, giving developers a foundation for building AI-assisted radiology reporting tools. (link)
Rivia automates clinical trial data checks with new module: Data Findings flags issues for study teams to review, with early adopter Scancell expecting to shorten end-of-study data cleaning from 8 weeks to 2. (link)
Mosai combines AI and clinical review for home health assessments: OASIS Intelligence checks assessments against referral documents and clinical notes, with certified clinicians validating discrepancies that could affect coding, reimbursement, and audit readiness. (link)
Grok Bot adds doctor research and appointment booking: Elon Musk says the agent can identify nearby doctors, call their offices, and schedule appointments, extending its capabilities into patient access and healthcare navigation. (link)
Google DeepMind adds drug discovery tools to Antigravity: Researchers can use an AI agent to gather chemical data, predict compound potency, and propose molecules for laboratory testing. A company demonstration completed a lung cancer drug-design workflow in about an hour. (link)
Anthropic upgrades Claude models for biomedical research and technical work: Fable 5.2 and Mythos 5.2 offer improved reasoning and up to 30% faster output, the company says. Biomedical researchers can apply for Mythos access through its Life Sciences Verification Program. (link)
HealthRex releases realistic EHR environment for testing AI agents: The open-source SimEHR platform requires agents to navigate charts, review results, place orders, and write notes through a visual interface, supporting the upcoming PhysicianBench-UI benchmark. (link)
Jefferson shifts toward building more of its own AI applications: The health system plans to emphasize internal development and partnerships rather than primarily purchasing finished tools, seeking applications more closely aligned with its workflows. (link)
Partnerships & Deployments 🤝
Deerfield + AWS: AWS will serve as a preferred AI provider, giving selected portfolio companies cloud services, training, and technical support to develop AI applications across drug discovery, administrative workflows, and care delivery. (link)
Chai Discovery + Roche: Roche will apply Chai’s AI models and platform across drug discovery programs targeting multiple therapeutic areas, extending the platform’s use in pharmaceutical research. (link)
HCA Healthcare + Google: Their AI-supported nurse handoff tool is expanding to approximately 100 hospitals over the next 2 months following pilots in 2 divisions, organizing patient information for shift-change reports. (link)
Neurophet + RapidAI: Neurophet will integrate its brain imaging AI into RapidAI’s platform to support treatment monitoring for Alzheimer’s disease and multiple sclerosis, expanding access through existing imaging workflows. (link)
AZmed + RapidAI: AZmed’s FDA-cleared AZtrauma and AZchest tools are available through RapidAI’s enterprise platform, connecting AI-supported fracture and chest X-ray findings with radiology, emergency, and orthopedic workflows. (link)
Synthpop + Microsoft: Synthpop is scaling its AI agents for intake, patient engagement, insurance coverage, and payment on Azure Kubernetes Service, with support from Microsoft’s startup program. (link)
Ember AI + Bone and Joint Institute of Tennessee: The practice reports an 80% reduction in cost per coded surgical case using Ember alongside athenaOne, with company-reported accuracy comparable to an expert coder. (link)
Research & Evidence 🔬
Researchers silently steer AI depression treatment recommendations: In simulated cases using 1 open model, hidden changes during response generation shifted the emphasis between medication and self-care without changing its weights or system prompt. (link)
AI heart measurements modestly improve survival-risk prediction: Among 27,382 patients, AI measurements of heart pumping function predicted mortality slightly better than clinician measurements, although researchers emphasized that human oversight remains essential. (link)
Pre-visit AI summaries shorten consultations by 7 minutes: In a randomized trial of 300 neurosurgery patients, plain-language summaries improved patient-rated involvement and shortened visits, although physicians found the consultations somewhat more demanding. (link)
Retinal images help AI estimate personalized mortality risk: RetiME models lifespan trajectories from retinal photographs, with validation across European, Asian, and North American cohorts supporting further research into eye scans as broader health indicators. (link)
Google tests hypertension screening using everyday smartwatch sensors: A 196-person study evaluated AI that combines passive sensor data over multiple days to identify hypertension without cuff calibration, using ambulatory blood pressure monitoring as the reference. (link)
Lindus reports Phase 2a trial completed for under $1M: The company says its AI-powered Citrus platform supported a randomized ME/CFS trial at roughly one-thirteenth the total cost of comparable studies. (link)
Jefferson’s heaviest documenters see larger ambient AI gains: In a rollout to 5,234 clinicians, regular Abridge users spent 10–17% less time in notes. The heaviest documenters cut after-hours work by nearly 17 minutes daily. (link)
Policy & Regulation 🏛️
Hospitals weigh blocking clinician chatbots or buying access: Monument Health restricted OpenEvidence over compliance concerns, highlighting tensions between clinician adoption and institutional requirements for privacy, security, and authorized patient-data use. (link)
115+ Illinois hospitals rehearse AI-amplified cyberattack response: An Illinois Health and Hospital Association exercise simulated sophisticated cyber disruptions, testing coordination and emergency preparedness as attackers incorporate AI into their operations. (link)
Weill Cornell discloses unauthorized patient-data use in AI tools: Radiology patient information was entered into AI tools during a June research project against institutional policy. The organization suspended the research. (link)
White House tightens screening requirements for synthetic DNA: Federally funded researchers must transition to certified suppliers within 250 days for synthetic nucleic acids and 340 days for benchtop synthesis equipment. (link)
Xsolis breach affects 9 health systems and 1.4M patients: Updated reporting identifies additional organizations affected by the previously disclosed incident involving the utilization-management AI company, which says it is unaware of any misuse of the exposed information. (link)
Deal Desk 💰
Spotlight on latest capital raises, M&A, and investments…
FUNDING
OpenAI, a frontier AI lab, reportedly closed about $11B in new funding at a $1.25T valuation, up from its valuation of $852B in March. (link)
Isomorphic Labs, an AI drug discovery company spun out of Google DeepMind, is in talks to raise about $5B at a $50B valuation. The round has not closed and terms could change. (link)
Formation Bio, a New York-based AI-native pharma company, announced a $600M debut fund to acquire drug programs. Limited partners include foundations, endowments, and hospitals. (link)
Laza Medical, an AI-powered cardiac imaging company, raised up to $160M in Series B financing led by GE HealthCare to advance its robotic imaging system for cardiac procedures. (link)
Kalepa, a New York-based AI company for insurance, raised $75M in Series B funding at a $470M valuation. Updata Partners led, joined by IA Ventures and Inspired Capital. (link)
nPhase, an AI-native clinical data company, raised $30M in Series D funding led by Viking Global Investors to expand its platform for clinical research. (link)
Siggy, a New York-based AI mental health service, raised $6M in pre-seed funding. Village Global led, joined by PsyMed Ventures and Mystic Ventures. (link)
M&A
Ellipsis Health + Kouper Health: Ellipsis acquired the AI care navigation company to combine its voice agent with Kouper’s EHR integrations and longitudinal patient data. (link)
LeanTaaS + LookDeep Health: LeanTaaS acquired LookDeep to add AI-powered patient monitoring and computer vision to its iQueue hospital operations platform. (link)
Equiti + Lexi AI: The acquisition brings Lexi’s healthcare AI language support together with Martti’s qualified medical interpreters, expanding the options available to organizations serving patients with language-access needs. (link)
Other Relevant News 🔍
News, podcasts, blogs, tweets, resources, etc…
AI Job Opportunities 💼
Explore our AI Job Board or contact us to feature roles in our newsletter…
Lead Product Data Analyst at Hippocratic AI, AI clinical agents for healthcare
N/A | Menlo Park, CA (link)
Head of Engineering at Compiled Health, clinical AI research company
N/A | Remote, North America (link)
Clinical Sales Lead at AssistIQ, AI supply and implant tracking for hospitals
N/A | Remote (link)
Visuals of the Week 📸
Funny memes, cool pics, and interesting data from around the web…
That’s it for this week friends! Back to reading — I’ll see you next week.
Stay classy,
PS. I write this newsletter for you. So if you have any suggestions or questions, feel free to reply to this email and let me know
How was this week's newsletter? Tap your choice below👇
