← All health signals
Consumer behaviour

The First Opinion

About a third of US adults ask chatbots about symptoms and lab results. Many never follow up: 42% on physical health, 58% on mental health.

What is it?

People take their symptoms, lab results and diet questions to a chatbot, and for many the answer is where it ends. A third of American adults now ask one about health, and in one poll most who asked about mental health never followed up with a professional.

Why now?

Three national surveys published in 2026 found a quarter to a third of adults asking chatbots about health. The young lead, at 44% of those under 30. In the UK, 15% of the public say they used a chatbot instead of contacting a GP.

Who’s building it?

Adults under 30, people checking diet and training, and anyone waiting, paying or unsure whether to book.

Young adultsDiet checkersSymptom searchersWaiting patientsGPs
The full read
Why now, who is building it, where the market goes, and what stays unsettled.
Get weekly →
01 · The market
Who’s in it
Builders — chatbot makers whose general tools now answer health questions.
Buyers — adults asking about symptoms, results and diet, often before or instead of booking.
Gatekeepers — clinicians and health services who see only those who follow up.
What’s at stake
Whether a free private answer becomes the end of the health question, or one step towards care.
Whether the skipped visits were needed at all.
Who notices when a chatbot answer is wrong.
What’s unsettled
Whether people who skip follow-up needed care in the first place.
Why mental-health askers follow up far less than others.
How far the habit reaches beyond the two countries where it is measured.
02 · How the demand matureda progression, not a timeline
Novelty
A chatbot answers a symptom question at once, privately and for free.
Retention
Diet, training and lab results bring users back between and around visits.
Willingness to pay
Cost and access push some towards the free answer, and some skip a paid visit.
Category
A health question can now start and end in a chat window.
03 · Where the market goesthe forks, not a prediction
→The companion reading: the chatbot stays beside the doctor, used before and after visits, and follow-up gaps narrow as people learn its limits.
→Mental-health questions settle in the chat window, with few ever reaching a professional.
→Health services start meeting fewer early problems, and more that arrive late, after a private answer.
→Trust stays low while use keeps widening, and people act on answers they doubt.
The forks turn on whether an answer that feels complete tells people when a question needs a clinician, and whether they listen when it does.
04 · The Health Deskadjacent market, moving now
The Pre-Diagnosed Patient
Patients arrive at the clinic already researched.
Over 60% of dermatology patients now search online before a visit.
JAAD International
Machine Therapy
Emotional questions move to a tireless listener.
27% of U.S. adults have used AI chatbots for personal, emotional or social queries.
The Washington Post
The Unlicensed Diagnosis
Regulators draw the line around consumer health tools.
FDA’s wellness guidance exempts a product only while it stays unrelated to diagnosing a disease.
FDA.gov
05 · The Wire
01
34% of US adults use AI chatbots for health reasons.
Pew Research Center ↗
02
58% of mental-health askers never followed up with a professional.
KFF ↗
03
About 14 million US adults skipped a provider visit after using AI, and only 4% of users strongly trust it.
Gallup ↗
04
15% of the UK public have used a chatbot for health advice instead of contacting a GP or NHS service.
King's College London ↗
The panel on this signal
They agreed on
–No survey measures whether the chatbot comes before a doctor, so the signal rests on follow-up, not order.
–Most users treat the chatbot as a companion to care.
–The follow-up gap is widest for mental health, where most askers never reach a professional.
The telephone-triage nurse disagreed on
–Most health questions resolve without a clinician; skipped follow-up matters only when a chatbot misses the red flags triage catches.
Know more about the method →
✦How this signal was built
01 · Panel
A Delphi panel of 4 AI analysts, each reasoning independently.
→
02 · Debate
They challenge and revise each other’s reads over two rounds.
→
03 · Evidence
Every claim is tied to a cited, linked source on The Wire.
→
04 · Synthesis
A moderator writes it into one read, disagreement and all.