
Stanford & Harvard trained MD-Scientist. Founder, https://teapawt.net/t.co/1eE8zLQbhz & https://teapawt.net/t.co/YGpenFxGpu, Partner @Continuum_VC, Adjunct Prof @StanfordMed
Stanford, CA
Joined November 2008
- Tweets19.2K
- Following8.1K
- Followers56.8K
- Likes28.4K
Pinned Tweet
Healthcare is shifting from a place you go to a system that surrounds you.
As we turn the page from 2025 into 2026 and beyond, we’re moving past simply digitizing yesterday’s workflows and into an era of real reinvention- where data, AI, and new care models converge and start to change the operating system of health itself.
Here is the arc of the possible as we move from NOW → NEAR → NEXT
•Sick Care → Health Care → Self-Health/Health Creation
Moving from reacting to disease to actively building capability, literacy, and tools so people can manage more of their health day-to-day.
•Episodic → Continuous → Predictive / Preemptive
From snapshots to signals: where we catch trajectory shifts early and intervene before “diagnosis day.”
•Physical → Virtual → Phygital + Ambient Care
Care won’t be confined to the clinic. Hybrid becomes default, and “ambient” care fades into the background—always-on, low-friction, context aware.
•One Size Fits All → Personalized → Adaptive (Real-Time) Precision
Personalization is step one. The leap is dynamic care plans that adjust in real time—dose, behavior, coaching, follow-up—based on data + context.
•Provider → Crowdsourced → Collective Intelligence (Networked Care)
Expertise becomes distributed: clinicians + communities + protocols + data networks learning together, not in silos.
•Human Cognition → AI-Enhanced → AI + Agentic Orchestrated (Copilot Teams)
We’re moving beyond “AI that writes notes” to copilots that coordinate care loops—triage, follow-up, prior auth, scheduling, escalation—freeing humans for judgment, empathy, and trust.
•Fee for Service → Fee for Outcome → Fee for Healthspan
The long game is aligning incentives around sustained function, prevention, and longevity—shared value for keeping people well.
My take: in 2026, the bottleneck won’t be model capability. It’ll be workflow integration, trust, governance, and getting incentives aligned (not to mention biomedicine funding challenges and improving trust in science).
More @NextMedHealth
Great to be in London & to share a session with Hannah Fry @FryRsquared at @UBS Biopharma conference- exploring the future of AI enabled health and biomedicine.
Incredible advances in Brain Computer Interfaces (BCI) converging with AI— enabling the disabled
Amazing work from @neuroleigh , @BrainGateTeam @MassGenBrigham et al
nytimes.com/paidpost/mass-ge…
Daniel Kraft, MD retweeted
How multimodal AI, combined with blood test omics via "liquid biopsies," will transform cancer monitoring and enable earliest detection.
open-access @CellCellPress
cell.com/cell/fulltext/S0092…
The genome is going to become increasingly actionable across health and medicine
Super proud to announce AlphaGenome 2 - our new DNA sequence model! 🧬 It predicts the impact of DNA variants in different specific cell types - unlocking crucial information for understanding complex diseases. Available for academic research via our API: alphagenomedocs.com/
🚀
Excited for the launch of Gemini for Healthcare!
We’re extending the Gemini agent with specialized healthcare skills and connectors to Epic, Optum, and more, built to support HIPAA compliance.
Instead of spending hours piecing together fragmented records and policies, care teams can offload the admin burden and get time back for what matters most: patient care.
3
1,170
Drones to the rescue.
The battlefield has been a catalyst for many innovations in medicine.
Korea and Vietnam gave us the medevac helicopter.
The Ukraine war may give us the flying ambulance.
wapo.st/3WPbDjl
Daniel Kraft, MD retweeted
Over 1 billion people live with hypertension, yet more than half don’t know they have it because traditional screening relies on periodic clinic visits or cumbersome cuff measurements.
In our @Nature publication, together with @BIDMChealth, and @harvardmed we present a multimodal artificial intelligence system capable of passively detecting signs of hypertension using a consumer smartwatch without requiring any cuff calibration.
Validated against the clinical gold standard of 24-hour ambulatory blood pressure monitoring (ABPM) across five U.S. sites, the system achieves accuracy levels comparable to traditional cuff-based diagnostic workflows for both initial and confirmatory screening.👇
What if your annual physical became a continuous physical?
Wearables + AI could turn everyday health data into insights your care team can act on, between visits.
An 80-second glimpse from my conversation with Moira Gunn on @BiotechNation 👇
Listen to our full conversation BioTech Nation: “Are we moving from Dr. Google to Dr. ChatGPT?”
We explore how AI is reshaping health and medicine—and what comes next.
biotechnation.com/episodes/a…
1
2
309
Gutting American Science... from DARPA to the NIH and beyond. Incalculable harm done.
nytimes.com/2026/10/31/scien…
Daniel Kraft, MD retweeted
Quite a lede
"For the first time, scientists say they have used viruses designed by AI to treat a human patient."
Front page @nytimes
gift link nytimes.com/2026/10/20/scien…
Sequential diagnosis with language models | Science
science.org/eprint/Y7CZQFPZM…
The Medical Selfie: Facial appearance offers a non-invasive window into health…
There's a lot in a face photograph about health that AI can detect, as seen in this foundation model of >40 million face images, including outcomes from cancer, biological age, influence of lifestyle factors, and survival @LancetDigitalH
thelancet.com/journals/landi…
Daniel Kraft, MD retweeted
It's late 2026 and we still don't have a way to assess a person's immune system status in the clinic, the underpinning of healthy aging. But soon there may be at least 3 ways!
erictopol.substack.com/p/a-t…
A truly amazing night with @JacobCollier at the @Castro_Theatre in SF.
Being part of his “Audience Choir,” a packed house harmonizing in real time under his direction, was pure joy.
A once-in-a-generation musical genius. If you haven’t discovered him yet, you’re in for a treat.
3
5
1,125
Daniel Kraft, MD retweeted
Is it over for human doctors?
No. Airliners can largely take off, fly, and land themselves. We still put two pilots up front, for judgment, edge cases, and trust. Same in the exam room.
My take on Washington's AI-in-medicine bet, on @NewsNation with @jessecordweber
Daniel Kraft, MD retweeted
凄い!!!
「レントゲンの骨や臓器をAIが自動で色分け」
・ MITらのFleXray、60種類の解剖構造を識別
・ 既存CTから正解付きの模擬X線画像を作り学習
・ さまざまな撮影部位や角度に対応。研究用デモも公開
・ 身近なX線検査から、骨の変形の自動計測や治療支援へ
flexray.csail.mit.edu/

















