Three ages. A photo shows only one.
Your ID card, your face and your cells can tell three different stories. An age test on a photo reads only the face, and only on that photo.
Updated 2026-10-18
Chronological age: the date on your ID
The time since you were born. It is a fact, not an estimate, and no photo can measure it. Any service that needs your real age (alcohol, online access, contracts) needs a document or an approved verification method, not a fun test.
Apparent age: what your face suggests
The age other people would guess from your face. It moves with light, expression, hairstyle, beard, make-up, tiredness and the photo itself. This is what Face-Age estimates: the FairFace model was trained on perceived ages, where several annotators judged the age group of each face.
Biological age: a research measure
Researchers use “biological age” for estimates of how far the body’s ageing processes have gone, built from biological markers rather than from a calendar. A well-known example is the DNA methylation clock published by Steve Horvath in 2013, built from about 8,000 tissue samples. These measures come from laboratory analyses. A selfie cannot provide them.
Are they linked?
Sometimes, at population level. In a Danish study of 1,826 twins aged 70 and over, published in the BMJ in 2009, the age assessors gave from photos was associated with survival, even after taking real age into account. Within twin pairs, the one who looked older was more likely to die first.
That result concerns people over 70, standardised photos and human assessors, analysed as a group. It does not turn an individual estimate from an app into a health indicator. Face-Age makes no claim about health, ageing speed or the effect of a product.
What to take from your result
“30-39 on this photo” means: on this picture, with this light and this expression, the model finds you closest to faces annotators placed in the 30-39 group. Another photo can say something else, and that is normal. To understand why, read why two photos give two results.
Sources
- Christensen et al., Perceived age as clinically useful biomarker of ageing: cohort study, BMJ 2009;339:b5262.
- Horvath, DNA methylation age of human tissues and cell types, Genome Biology 2013;14:R115.
- Kärkkäinen and Joo, FairFace: Face Attribute Dataset for Balanced Race, Gender, and Age, arXiv:1908.04913.