What it takes to trust an AI with your health
A Johns Hopkins AI researcher on what made her system stick
Levels <[email protected]>
Share
Levels promotes an episode of NextLevel featuring Dr. Suchi Saria, who discusses building trust in medical AI. She shares her experience developing a system that detects early signs of sepsis and explains how to get hospitals and clinicians to buy in. The episode highlights five key takeaways for trusting medical AI, including building trust with everyone and continuous data interpretation.
What the raw message shows
From the headers
- Sent with
-
SendGrid See it in the headers
Across Levels's email
- Sends
- About 3 a week
- DMARC
-
Enforced, reject
Mail failing authentication is refused.
The copy
400 words · about 2 min to read · 2 buttons
The new cholesterol numbers to know
The hardest part of medical AI is getting people to believe it
Plenty of AI tools can analyze medical data. Far fewer get used by the doctors and nurses they're built for, and fewer still show that patients actually do better. Dr. Suchi Saria, a Johns Hopkins computer scientist and founder of Bayesian Health, built one that did both: a system that watches hospital patients for early signs of sepsis, a dangerous overreaction to infection. It took more than a decade, and much of the hard work came after the algorithm existed. On the latest episode of NextLevel, she explains how she got hospitals, clinicians, and regulators to buy in, and what that means for continuous health monitoring.
Button How to trust medical AI →
Five Key Takeaways
1. Trust has to be built with everyone at once.
Hospitals need to know a tool is easy to install and will pay for itself, and clinicians need alerts that mean something and leave the final call to them. Everyone needs peer-reviewed evidence that it works.
2. Continuous data needs continuous interpretation.
Hospitals were already capturing a steady stream of vitals and labs but often reviewing it in batches hours apart. The value came from reading each new measurement as it arrived, against that patient's own normal.
3. The best alerts know when to stay quiet.
Clinicians are interrupted so often by simple, rule-based alarms that they learn to ignore them. Saria's system speaks up when it's confident and waits for more data when it isn't, and clinicians evaluated nearly 9 in 10 of its alerts, compared with the 10 to 15 percent typical of health record alerts.
4. AI earns responsibility the way a new hire does.
Saria's rule for high-stakes care is "verify, validate, and then trust," handing over more as a track record builds. Clinicians said their trust came from hands-on experience, colleagues' endorsements, and keeping control of the decision.
5. Medicine is aiming too low with AI.
Hospitals mostly use it for billing and paperwork today. Saria sees the bigger opportunity in catching preventable harm as it happens, since diagnostic errors are linked to an estimated 371,000 US deaths a year.
Button Read the full story →
Looking to understand more about your metabolic health?
Explore our health news and advice site, Metabolic Insights
228 Park Ave South, PMB 63877, New York, NY 10003
The words as text: what a screen reader can follow, and all there is when images don't load.
What the headers say
Read from the message's own headers, as it arrived.
- Came from
-
o33.sg.m.dripemail2.com
· 168.245.99.76
SendGrid's sending servers
- SPF
- Pass for drip.levels.com
- DKIM
-
- Pass signed by levels.com
- DMARC
-
Pass for levels.com
Policy: reject
- Unsubscribe
-
One-click
Meets the Gmail and Yahoo rule for bulk senders.
- Complaints
- No feedback-loop header
Show all 17 headers Hide the headers
As received, minus the delivery side: the archive's own addresses show as [recipient], its mail servers as [receiver], signatures are left out and unsubscribe links trimmed to their host.
Route
- Received
- from o33.sg.m.dripemail2.com (168.245.99.76) by [receiver] (cloudflare) id iPRSiAVzUBwz for <[recipient]>; Thu, 29 Oct 2026 16:00:20 +0000
- Received
- by recvd-5cffb6dd75-v4f56 with SMTP id recvd-5cffb6dd75-v4f56-1-6AE366B9-9A 2026-10-29 15:31:05.298439324 +0000 UTC m=+2586634.763266121
- Received
- from MTMxMTA3NDc (unknown) by geopod-ismtpd-24 (SG) with HTTP id rWRH_-pPSt-XwVxE2_Jqvw Thu, 29 Oct 2026 15:31:05.280 +0000 (UTC)
Authentication
- Received-SPF Authentication result
- pass ([receiver]: domain of [recipient]@drip.levels.com designates 168.245.99.76 as permitted sender) receiver=[receiver]; client-ip=168.245.99.76; envelope-from="[recipient]@drip.levels.com"; helo=o33.sg.m.dripemail2.com;
- Authentication-Results Authentication result
- [receiver]; dkim=pass header.d=levels.com header.s=s1 header.b=dzyrxeNj; dmarc=pass header.from=levels.com policy.dmarc=reject; spf=none ([receiver]: no SPF records found for [email protected]) smtp.helo=o33.sg.m.dripemail2.com; spf=pass ([receiver]: domain of [recipient]@drip.levels.com designates 168.245.99.76 as permitted sender) smtp.mailfrom=[recipient]@drip.levels.com; arc=none smtp.remote-ip=168.245.99.76
Platform
- X-Drip-SendingAccount
- 4080355
- X-SG-EID Identifies SendGrid
- u001.79kJZC85z/WE9t9nOmpiHosf7qU1Au+mrKy3jytIn8sMbSWRTgBdv9q1OykY6JKW2YJWJHg2PyWC+wsUdQDHFai6sTkOKrEJrJgIg0KK8dfLoEt7+MbHANRlMF+EYUz+SC/iyy6tFwdoiw0IQV/eSxK6wshbLbbZPVOUFFtaSyF1TK0ItHS42wZNue31Wv6VGwDszVqquRwBdLda575b20lfGEbld/sg5pvDWoKIz30=
- X-SG-ID Identifies SendGrid
- u001.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
- X-Entity-ID
- u001.dEl2Z4sQ4dqI57TjhSSBBA==
Unsubscribe and feedback
- List-Unsubscribe
- <mailto:…@unsubscribe.dripemail3.com>, <https://unsubscribe.dripemail3.com/…>
- List-Unsubscribe-Post
- List-Unsubscribe=One-Click
Message
- Content-Type
- multipart/alternative; boundary=f4a097d932311c288d01e4dca270be822a6df7529c1b1d40f2ef98d492df
- Date
- Thu, 29 Oct 2026 16:00:00 +0000 (UTC)
- From
- Levels <[email protected]>
- Mime-Version
- 1.0
- Message-ID
- <rWRH_-pPSt-XwVxE2_Jqvw@geopod-ismtpd-24>
- Subject
- What it takes to trust an AI with your health
More from Levels
All 29 emailsSimilar campaigns from other brands
Different brands running the same play. Compare any of them side by side.
Archived for reference and commentary. Levels, its logos and content are the property of their respective owners; Inbox Archive claims no affiliation and has disabled all links. If you own this content and would like it removed, get in touch.