Since autumn 2025, heart rate can be measured not only by watches and fitness bands but by earbuds too: AirPods Pro 3 have a built-in optical sensor that shines into the ear canal. For a runner this sounds tempting — fewer devices, and an ear, unlike a wrist, barely moves around. But how far can you trust these data? The first independent test was published in spring 2026.

What the researchers did

Doherty's team at University College Dublin (Ireland) (PLOS Digital Health, April 2026) carried out a laboratory validation.

40 adults, average age 23.8, 37.5% women.
A treadmill protocol: rest, then stages from walking to running at roughly 40 to 85% of predicted maximum heart rate, including abrupt speed changes.
The reference was a Polar H10 chest strap, which measures heart rate from the heart's electrical signal (like an ECG).
Data from both devices were time-synchronised and split into 5-second segments — 16 735 pairs of readings in total.
The earbuds were bought at an ordinary retailer, and Apple was not involved in the study. Funding came from Irish government research agencies.

What they found

There is practically no systematic bias: on average AirPods read 0.03 beats lower than the Polar.
The mean absolute error was 2.08 beats per minute, or 2.02%.
Across intensity zones, the average error stayed within 1.3–2.4 beats — from rest to hard exercise.
But the spread of individual readings is noticeable: for 95% of five-second segments the difference from the reference was within ±10 beats, and the higher the heart rate, the wider this band.

For comparison, the authors cite a meta-analysis of the Apple Watch: there too the average bias is less than one beat, and the error band is about ±7 beats. So on average the earbuds are no worse than a watch, but are a little “noisier” at high intensity.

Why the ear and not the wrist

A sensor in the ear is closer to the central circulation and suffers less from blood vessel constriction in the cold than the wrist. Even more important for runners: wrist optics sometimes mistake step rate for heart rate — the cadence lock effect, covered in detail in a separate article. The ear has its own sources of interference: head and jaw movement and a poorly fitting ear tip. The authors attribute the growing spread at high intensity precisely to these.

Limitations

Lab and treadmill. No wind, cold, rain or sweat pouring into your ear in the third hour.
Peak effort was about 85% of predicted maximum heart rate. Short maximal intervals and sprints were not tested.
Young, healthy participants, with Fitzpatrick skin types I–IV only. Accuracy may differ in older people and on darker skin.
One firmware and one iOS version: algorithm updates could make results either better or worse.
Ear tips were fitted by lab staff following the manufacturer's instructions — in real life, the fit is often worse.

How to apply this

Easy runs, recovery, long runs. For keeping things “no higher than zone 2”, an average error of 2 beats is an excellent result. The earbuds are good enough.
Tempo, threshold, intervals. Here the boundary between neighbouring zones matters, and a ±10-beat band can tip you from one into the other. For these workouts a chest strap is more reliable.
Watch the trend, not the number. A single 8–10 beat spike at high intensity may be noise. Go by the average over the rep.
Fit is everything. Choose an ear tip size so the earbud doesn't shift while you run: the quality of the optical signal depends on how snug the contact is.
Don't use the earbuds to determine your maximum heart rate. For that you need a chest strap and the right protocol — on how to find your maximum, see our article on maximum heart rate formulas.
Zones based on your threshold or maximum are easy to recalculate in the zones generator.

The bottom line

The first independent test of AirPods Pro 3: 40 people, 16 735 paired readings against a Polar H10 chest strap.
No systematic bias, and an average error of about 2 beats (2%), on a par with the best watches.
At high intensity, individual readings diverge from the reference by ±10 beats.
For easy running and tracking trends the earbuds are enough; for intervals and precise zone boundaries, use a chest strap.

Sources: Doherty C., Burke D., Mitchell O., Lawless D., Brar J., Fuchs M., Lambe R. “Accuracy of heart rate measurement using AirPods Pro 3 during graded treadmill exercise: A laboratory-based validation study”. PLOS Digital Health, 2026;5(4):e0001373. DOI: 10.1371/journal.pdig.0001373