The future of health is prediction.

More insight from the health data we already have.

Humanity helps partners turn routine blood tests into biological age insights.

Today, biological age. Tomorrow, far more: cancer and heart risk, women’s health, and new treatments for disease.

Available through leading labs, including Quest Diagnostics, Limbach, ALAB, SPADIA and NRL.

Every blood test holds part of a bigger picture.

Most measure only a handful of markers.

The connections reveal more.

Our models use learned relationships between markers to estimate the ones that weren’t measured.

The bigger picture, from the same blood test.

Today, that means biological age insights, from a model trained on real health outcomes.

From research to the lab, in days.

AI agents keep testing new ways to train our models. Proven gains reach partner labs as new, validated versions.

In development

Learning cause and effect, not just correlation.

With researchers at leading AI institutions, we’re training models to learn how biology actually works, so they need even less data.

Tomorrow, far more than biological age.

For partners working on everything from cancer and heart risk to women’s health and new treatments for disease.

Our partners

Trusted by science. Chosen by industry.

Imperial College London University College London
Nature Communications Biology
Quest Diagnostics Limbach Gruppe ALAB Laboratoria
SPADIA Lab National Reference Laboratory, brought to you by M42 London Medical Laboratory EE

Secure and data compliant

SOC 2 Type 2 ISO 27001 certified GDPR compliant

Why labs partner with us

Works with the blood tests labs already run.

Customers get a clear read on their health long before they get sick. Labs see higher conversion, more repeat tests and deeper engagement.

  • No new test
  • No new sample
  • No new workflow

No patient identifying data ever leaves the lab.

The approach

We did not build a bigger telescope. We changed the sensor.

The data a real person has is thin and uneven. A bigger instrument does not fix thin data. A better sensor does.

A real person has only a few markers, from one test or from many.

We use imputation to predict the markers that are missing.

Then we read the whole constellation, and accurately predict a biological age.

This is only the start. The same method sharpens everything else we predict about health.

The proof

Peer reviewed, and ahead of the previous gold standard.

Published with Imperial College London in Communications Biology, a Nature Portfolio journal. It works from a common blood panel, like a complete blood count.

Concordance per panel Predictive concordance by blood panel. PhenoAge scores 0.731. Humanity Age scores 0.733 from a complete blood count, rising to 0.743 on expanded and comprehensive panels. BMI scores 0.712 and CMP 0.718, both below PhenoAge. PANEL CONCORDANCE Expanded + Comprehensive 0.743 Routine Health Panel 0.738 CBC + Cholesterol 0.735 Complete Blood Count 0.733 PhenoAge 0.731 CMP 0.718 BMI 0.712 PREVIOUS GOLD STANDARD Routine 20 to 30 markers. Expanded and comprehensive 40+. Concordance per panel Predictive concordance by blood panel. PhenoAge scores 0.731. Humanity Age scores 0.733 from a complete blood count, rising to 0.743 on expanded and comprehensive panels. BMI scores 0.712 and CMP 0.718, both below PhenoAge. CONCORDANCE Expanded + Comprehensive 0.743 Routine Health Panel 0.738 CBC + Cholesterol 0.735 Complete Blood Count 0.733 PhenoAge PREVIOUS GOLD STANDARD 0.731 CMP 0.718 BMI 0.712 Routine 20 to 30 markers. Expanded and comprehensive 40+.

Read the science →

In the lab

Prediction is all you need.

Biological age is the proof, not the product. We are building a large biological model that makes every health prediction sharper, not just this one.

What it will predict: who is at risk and how early, which treatment is most likely to work for whom, and what a change actually did.

Coming soon

Request a demo

Become a partner.
Join us on the mission.

A billion extra years of healthy life, given back to the world. Bring it to the people you already test.