/
Navigation
Chronicles
Browse all articles
Explore
Semantic exploration
Research
Entity momentum
Nexus
Correlations & relationships
Story Arc
Topic evolution
Drift Map
Semantic trajectory animation
Posts
Analysis & commentary
Pulse API
Tech news intelligence API
Browse
Entities
Companies, people, products, technologies
Domains
Browse by publication source
Handles
Browse by social media handle
Detection
Concept Search
Semantic similarity search
High Impact Stories
Top coverage by position
Sentiment Analysis
Positive/negative coverage
Anomaly Detection
Unusual coverage patterns
Analysis
Rivalry Report
Compare two entities head-to-head
Semantic Pivots
Narrative discontinuities
Crisis Response
Event recovery patterns
Connected
Search: /
Command: ⌘K
Embeddings: large
TEXXR

Chronicles

The story behind the story

days · browse · Enter similar · o open

About 15,000 US doctors and assistants are using an AI feature in MyChart, a ubiquitous communications platform in US hospitals, to draft replies to patients

Overwhelmed by queries, physicians are turning to artificial intelligence to correspond with patients.  Many have no clue that the replies are software-generated.

New York Times Teddy Rosenbluth

Context & Ripple Effects

Earlier coverage captured hospital tests of GPT-3 for online patient queries and clinicians using ChatGPT to make patient communications more empathetic. MyChart moves that use case from stand-alone experimentation into a widely used hospital communications channel.

The significance is distribution: AI drafting is being used by roughly 15,000 doctors and assistants where patient messaging already happens, rather than requiring clinicians to switch to a separate tool.

First-order effects

  • Doctors and assistants using MyChart can draft replies to incoming patient messages with AI, potentially reducing the writing burden created by high query volumes.
  • Patients may receive AI-drafted correspondence without knowing software contributed to the response, making clinician review and communication practices immediately consequential.

Second-order effects

  • Hospitals deploying AI in messaging will need to determine how drafts are reviewed and whether patients are told when software assists, because the tool operates in a patient-facing channel.
  • Clinical-AI vendors have a stronger incentive to embed communication features in established workflows, alongside the administrative automation seen in AI tools used to automate clinical notes.

Third-order effects

  • If adoption holds, patient messaging may become a standard AI-assisted administrative workflow, shifting value toward platforms that control the clinician’s daily communication surface.
  • The pattern points to a broader need for healthcare organizations to distinguish AI that prepares work from AI that independently communicates with patients; how they set that boundary remains unsettled.

The trend: Healthcare AI is moving from discrete clinician tools into workflow-native systems that draft and organize routine patient-facing work.

Discussion

  • @paleofuture.bsky.social Matt Novak on bluesky
    “One doctor received an email saying, ‘I want to be the first to congratulate you on your A.I. co-pilot and be the first to send you an A.I.-generated patient message.’” [embedded post]
  • @evankirstel @evankirstel on x
    That Message From Your Doctor? It May Have Been Drafted by A.I. https://www.nytimes.com/... [image]
  • @calonghurst Christopher A. Longhurst on x
    Many thanks to @teddyrosenbluth for this @nytimes story about 𝐀𝐈-𝐠𝐞 𝐧𝐞𝐫𝐚 𝐭𝐞𝐝 𝐝𝐫𝐚𝐟 𝐭 𝐫𝐞𝐩𝐥 𝐢𝐞𝐬 𝐭𝐨 𝐩𝐚𝐭𝐢 𝐞𝐧𝐭𝐬 I am grateful to be quoted about @UCSDHealth efforts along with esteemed colleagues across the country Full story⤵️ https://www.nytimes.com/... [image]