Unlike the traditional ways of working out your biological age, from solely blood tests or epigenetic data, recent developments in artificial intelligence now offer novel tools for clinicians to assess and potentially intervene in the ageing process.

Although an AI model can now complete an impressive number of tasks with better than human accuracy, we are still incredibly early on in the artificial intelligence age.

Face and Biological Age

FaceAge is an example of such technology that is meant to predict the biological age with up to 90% accuracy. It was trained on more than 58,000 individuals aged 60 years or older. It as developed by Harvard and Mass General Brigham and it uses a snapshot of someone’s face to predict the biological age and cancer survival time.

The FaceAge model has so far only been applied in studies to helping predict biological age in cancer patients, but I am sure in the near future we will see it applied to other groups of people too.

It’s only a start, but it goes to show that an AI model applied in the right way can be used to greatly save time and costs. Currently, validated biological age models such as PhenoAge and DunedInPace can cost upwards of £500. Perhaps using FaceAge instead, once the AI mode is accurately trained, could be used to bring the costs to <£1 per biological age reading.

This would have huge implications on the longevity speciality once an accurate biological age prediction method becomes more accessible.

The face however, is only a start!

The Eyes

Deep-learning systems are now scouring other easily captured images for tell-tale signs of biological wear and tear.  A growing body of work on “retinal age” shows that algorithms can read minute changes in the blood vessels at the back of the eye from a fundus photograph and predict a person’s systemic biological age, the risk of stroke, kidney disease and—most recently—women’s reproductive ageing.

This was described in a June 2025 Nature paper. It seems that the eye could provide a non invasive view into both women’s health and overall healthspan.

Our Speech

Speech is also very interesting. Our voice may also betray how fast we are ageing.

Voice-biomarker studies published this spring show that short smartphone recordings can predict biological age and cognitive status with clinically useful accuracy. This is the latest study that was published in The Lancet.

Perhaps in the future, we’ll get a health score after every phone conversation, as our AI devices listen in the background and prompt us of any changes in pattern. Recognising a pattern change may point towards a possible change in our health and lead to earlier interventions.

Until the AI ecosystem develops accurate enough models to be used in longevity centres, we have to rely on blood tests and epigenetic markers. Contact us to book a consultation with our longevity experts.