← All beauty signals
Consumer mood

The Previewed Procedure

An AI app renders what a patient’s face might look like after surgery, before any surgeon agrees. A surgeon calls it a visual estimate, not a modelled outcome.

What is it?

An app can already render what a patient’s face might look like after a cosmetic procedure, before any surgeon agrees to attempt it. A named surgeon calls the result a surface-level visual estimate, not a modelled outcome — and no clear rule yet governs the apps producing it.

Why now?

Dermatologists named ‘Snapchat dysmorphia’ in 2018 for patients chasing a filtered selfie; generative preview apps now automate that habit and sharpen it. A named surgeon warns the training data lacks the diversity to model real skin tones, anatomy, and healing patterns across the populations using it.

Who’s building it?

Consumer preview-app makers, plus the surgeons now spending consultation time correcting the expectation the app set.

Preview-app makersCosmetic surgeonsDermatologistsAd regulatorsDevice oversight (absent)
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
Preview-app makers — selling a plausibility no surgeon can match.
Surgeons — spending consultation time correcting an expectation a render set, not the clinic.
Patients — comparing a real outcome against a render built on impossible anatomy.
What’s at stake
A generated preview is replacing the filtered selfie as the reference point a patient brings into consultation.
No clinical sign-off is ever attached to that image.
The app itself sets the expectation, not the clinic.
What’s unsettled
Should a consumer preview app count as a medical device?
Does a private preview ever cross into a marketed claim?
Whether training data biased toward some skin tones ever gets corrected before harm.
02 · How the demand matureda progression, not a timeline
The filtered selfie
named in 2018, the older habit these apps now sharpen.
The generated preview
formalises that habit into a rendered, surgically impossible face.
A named warning
a Wits surgeon says the data can’t model real skin tones or healing.
No governing rule
the apps sit outside both advertising doctrine and medical-device oversight.
03 · Where the market goesthe forks, not a prediction
→If a preview ever reaches marketing, the before-and-after doctrine already has an answer; the app used privately in consultation still has none.
→Regulators could finally extend real medical-device oversight to these apps, or leave them entirely unregulated.
→Training data could widen to model real anatomy, or stay narrow and keep producing impossible faces.
→Or a surgeon’s own warning could simply become part of the standard consultation script.
Underneath every fork sits one plain gap: a render can set a patient’s expectation long before any regulator, surgeon, or advertising rule ever reviews it.
04 · The Preview Gap deskadjacent signals, same render
The Rendered Result
A generated result claim, still unproven.
UK guidance treats an AI ‘after’ image as a claim.
ASA (CAP)
The Invisible Filter
Two countries already require an altered-body filter label.
France fines undisclosed ad retouching up to €37,500 or more.
NPR
The Impossible Brief
AI-made faces arrive as the client’s brief.
A celebrity makeup artist says clients bring AI-generated faces as references.
Newsweek
05 · The Wire
01
A surgeon calls some previews ‘surgically impossible’.
IOL / Sunday Tribune ↗
02
Patients brought filtered selfies to consultations by 2018.
ScienceDaily ↗
03
UK ad regulator: AI cosmetic-effect images can be materially misleading
ASA (CAP) ↗
04
An ‘after’ image must show what’s generally achievable.
ASA (CAP) ↗
05
The advertiser answers for the AI tool’s output, not the tool.
ASA (CAP) ↗
06
15% of 59 surveyed UK surgeons met AI-edited face requests
ITV News ↗
The panel on this signal
They agreed on
–A named surgeon’s warning is real and specific — the regulatory gap around these preview apps is real too.
–The mechanism is old; the app just automates it.
–If a preview is ever used in marketing, existing before-and-after doctrine already reaches it.
The medical-device regulatory specialist disagreed on
–This isn’t an advertising problem waiting for a ruling — a preview tool should already be regulated as a device.
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.