← All beauty signals
Governance

The Unlicensed Diagnosis

A cosmetic skin-scan app and a clinical skin-lesion device sit on either side of one legal test: intended purpose. Regulators drew that line explicitly in 2026.

What is it?

A cosmetic skin-scan app and a clinical skin-lesion device are separated by one legal test: what the product actually claims to do. FDA’s 2026 wellness guidance and device reclassification, plus the EU’s parallel intended-purpose rule, now draw that line explicitly, on both sides of the Atlantic.

Why now?

2026 is when the line stopped being theoretical. FDA finalised its wellness safe-harbour guidance in January, reclassified clinical skin-lesion devices in March, and that reclassification took effect in April, explicitly excluding consumer apps — three regulatory actions in one year, on the exact boundary a cosmetic skin-scan tool sits beside.

Who’s building it?

Regulators write and enforce the boundary; skin-scan and diagnostic-device makers build the products that test where it falls.

FDAEU regulatorsSkin-scan vendorsDevice makersDermatologists
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
Regulators — FDA and the EU, drawing the 2026 line.
Vendors — skin-scan and diagnostic-device makers whose claims decide which regime applies to them.
Physicians — the dermatologists who use the newly reclassified clinical devices.
What’s at stake
Which side of the line a tool’s own claim puts it on decides its whole regulatory burden, on both continents.
No named cosmetic brand has been tested against either line.
A wrong claim moves a product into the highest tier.
What’s unsettled
Whether any 2026 enforcement action will actually target a named cosmetic product.
Whether EU and FDA boundaries stay aligned as more tools launch.
Whether a screening claim can drift into a diagnostic one without anyone noticing.
02 · How the demand matureda progression, not a timeline
Wording
A tool’s marketing claim, not its technology, decides which regulatory regime applies.
Guidance
FDA states the cosmetic exclusion in writing for the first time this clearly.
Reclassification
Clinical devices move tiers while regulators state, in the same order, that laypeople’s apps stay untouched.
Test case
The architecture is drawn; enforcement against a named cosmetic product has not arrived.
03 · Where the market goesthe forks, not a prediction
→Whether 2026 produces the first enforcement action against a named cosmetic skin-scan brand, turning an architectural boundary into a tested one.
→Whether EU and FDA definitions of “intended purpose” stay aligned as more AI skin tools launch.
→Whether a screening tool’s claim can drift toward diagnosis under commercial pressure without regulators noticing quickly.
→Whether the EU’s AI Act adds a second layer on top of the device-classification question.
The line is drawn clearly on both continents in 2026 — whether any real cosmetic product gets tested against it is still unknown.
04 · The Intended-Purpose deskone test, two continents
The Safe Harbour’s Edge
FDA states the cosmetic exclusion plainly.
General Wellness guidance finalised January 2026, excludes cosmetics.
FDA.gov
The Other End Moves
Clinical skin-lesion devices reclassified, consumer apps stay excluded.
Class III to Class II, effective April 2026, not for consumer apps.
Federal Register
The Diagnosis Trigger
Wellness status ends when a claim turns to diagnosis or treatment.
FDA’s wellness exemption stops once a claim turns to diagnosing or treating disease.
FDA.gov
05 · The Wire
01
FDA’s wellness guidance excludes products that diagnose disease.
FDA.gov ↗
02
FDA moves skin-lesion devices to Class II; lay-user apps excluded
Federal Register ↗
The panel on this signal
They agreed on
–Regulators drew the cosmetic/medical-device boundary explicitly, on both continents, within months of each other in 2026.
–No 2026 enforcement action against a named cosmetic product exists.
–A tool’s stated claim, not its underlying technology, decides which regime applies to it.
The dermatologist disagreed on
–The architecture reads settled, but a tool that stays inside the safe harbour by wording can still miss a lesion.
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.