biological age

Biological age is a measure of how old your cells and organs actually are. Think of a car: it's not the odometer reading, but more like a mechanic's judgment. The odometer just tells you how many years and kilometers the car has been driven, but a mechanic listens to the engine, checks tire wear and oil condition, and separately judges how worn the car really is, regardless of its paperwork age. In the same way, how many years a person has been alive and how worn out their body actually is can be two different things.

Increasingly, AI is making this judgment instead of a person. Thousands of protein levels are measured from a blood sample, and when that pattern is fed into an AI model trained on such data, it outputs a number for how old this person's cells and organs actually behave. These predictive models are called aging clocks, and since each one is built a bit differently, the same person's blood can produce different numbers depending on which clock it's run through.

So when researchers want to check whether a new drug or treatment slows aging, they look at whether a patient's biological age went down before and after treatment. If multiple clocks point in the same direction at once, that's taken as evidence it's not just a coincidence.

In articles, you'll see phrasing like "after taking the drug, biological age dropped by about three years across all six aging clocks." What's easy to misread here is that this doesn't mean time was literally reversed and the person's calendar age went down. It means the predicted value an AI model calculates from blood protein levels changed, and whether this prediction actually translates into a longer lifespan still needs separate verification.

This isn't a concept you can manipulate hands-on, but asking an AI chatbot the following can help you get a feel for it.

Explain why biological age and actual age can differ, using the analogy of a car's odometer versus a mechanic's diagnosis, in simple terms.