SlipperSkin For app makers
Aisle · two guides

AI skin analysis

Selfie-based skin scanners and routine builders, read on their makers' own pages and set against independent tests. The claims that matter here are what is measured, what accuracy figure is published and against what, how it performs across skin tones, and what happens to your face.

The method, for this aisle

Five lines we check on every skin analysis app

Every guide here reads these lines on the maker's own page and dates what it found. The full method is on how we check.

  1. 01

    What it measures

    Wrinkles, spots, redness, pores or a "skin age", and from which photo.

  2. 02

    Accuracy published

    The figure, what it is measured against, and whether it is repeatability or accuracy.

  3. 03

    Skin tones

    Whether results are published by Fitzpatrick type.

  4. 04

    What it sells

    Whether it recommends its own or a client brand's products.

  5. 05

    Your face

    Whether the selfie is kept, for how long, and whether it trains the model.

Where the law stands. Where a tool scans a face, Illinois' Biometric Information Privacy Act defines a scan of face geometry as a biometric identifier, which is why these guides read what each app says it keeps.

☆ Save
Dashboard