Smartwatches nonetheless can not reliably measure blood sugar from the wrist. However Apple and Samsung could not want a breakthrough in sensors to make glucose monitoring a part of their subsequent AI well being battle.
As steady glucose screens (CGMs) transfer additional into mainstream wellness, the subsequent wearable contest could also be much less about detecting glucose instantly and extra about explaining it. Apple, Samsung, Oura, Dexcom, Abbott, and different health-tech gamers are circling the identical alternative: turning information on glucose, sleep, stress, train, diet, and coronary heart fee into personalised AI steerage.
That shift might make glucose monitoring the sharpest check but for client well being AI. The promise is helpful: real-time teaching. The danger is {that a} wellness nudge can begin sounding loads like medical recommendation.
Smartwatches nonetheless have a glucose drawback
The dream is straightforward: look at your wrist and see your blood sugar. The fact is just not.
Regardless of years of rumors and analysis, immediately’s main smartwatches and good rings don’t supply FDA-cleared, noninvasive blood glucose measurement. In 2024, the US Meals and Drug Administration warned customers to not depend on smartwatches or rings that declare to measure blood glucose with out pores and skin penetration, citing the danger of inaccurate readings and probably harmful well being selections.
That actuality creates a problem for Apple and Samsung. Glucose is among the most dear well being metrics they may add to their ecosystems, however it is usually some of the medically delicate. As TechRepublic not too long ago reported, corporations together with Apple, Samsung, Garmin, and Oura proceed to discover glucose-related well being options, at the same time as really noninvasive glucose sensing stays elusive.
So the near-term path will not be a watch that replaces a steady glucose monitor. It could be a phone-and-watch ecosystem that connects to CGMs, imports glucose information, and makes use of AI to elucidate what the numbers imply.
The AI layer could matter greater than the sensor
Steady glucose screens already generate streams of knowledge. For individuals with diabetes, that info will be medically important. For wellness customers, it could possibly reveal how meals, sleep, stress, and exercises have an effect on vitality and metabolic patterns.
However uncooked glucose curves will not be all the time simple to grasp. A spike after lunch, a late-night dip, or a sluggish restoration after a poor night time of sleep could also be helpful provided that the system can place it in context.
That’s the place AI turns into the true product.
As an alternative of merely exhibiting a chart, an AI well being assistant might summarize patterns, flag uncommon adjustments, join glucose swings to meals or exercises, and counsel inquiries to ask a clinician. The gadget on the wrist would nonetheless matter, however the aggressive benefit would transfer to the software program layer that interprets the physique’s alerts.
Apple and Samsung have already got the platform items
Apple and Samsung have most of the items wanted to compete in that path, even when neither firm has introduced a client smartwatch that may instantly measure blood glucose and not using a CGM.
Apple has the Apple Watch, iPhone, Well being app, HealthKit, and Apple Intelligence. Samsung has Galaxy Watch, Galaxy Ring, Samsung Well being, and Galaxy AI. Samsung has additionally been pushing towards extra AI-driven well being interpretation, with latest studies describing a redesigned Samsung Well being app constructed round personalised wellness insights, AI-generated summaries, Vitals, Vitality Rating, coronary heart well being, cardio load, and health metrics.
Apple’s present AI-health story is much less instantly tied to glucose. Its latest Apple Intelligence-connected health options, together with Exercise Buddy, level to a broader curiosity in utilizing AI to personalize well being and exercise steerage, however Apple has not introduced glucose-specific AI recommendation for customers.
That makes glucose information a potential subsequent frontier slightly than a confirmed product roadmap. A consumer’s glucose response to breakfast turns into extra significant when paired with sleep high quality, train, coronary heart fee variability, treatment logs, and stress patterns. AI can sew these alerts right into a story.
That can be why the competitors is about greater than wearables. It’s about who turns into the trusted interpreter of private well being information.
Maybe the clearest instance of that future is already right here.
In Could 2025, Oura expanded its partnership with Dexcom to deliver glucose insights from Dexcom’s Stelo steady glucose monitor into the Oura expertise. As The Verge reported, the characteristic combines glucose readings with Oura’s sleep, exercise, stress, and restoration metrics, then makes use of AI-generated summaries to assist customers interpret metabolic patterns.
There are necessary limits. The characteristic depends on Dexcom’s Stelo CGM slightly than glucose sensing from the Oura Ring itself, and The Verge reported that it was out there within the US at launch, with Stelo offered by means of Oura for $99.
That mannequin is necessary as a result of it separates sensing from interpretation. The CGM handles the glucose sign. Oura focuses on the AI-powered expertise round it.
In different phrases, the {hardware} gathers the info, however the AI creates the worth.
Apple and Samsung might comply with an analogous technique on a a lot bigger scale. They don’t essentially have to win the primary battle by constructing the proper noninvasive glucose sensor. They’ll win by making glucose information comprehensible, helpful, and habit-forming.
The FDA line is the arduous half
The problem is that glucose is just not a step depend. It isn’t a sleep rating. It’s a metric individuals could use to make selections about meals, train, treatment, and medical care.
That creates a slender path for client expertise corporations. An AI assistant can summarize developments, encourage customers to seek the advice of a doctor, or present academic wellness insights. However the nearer it will get to analysis, remedy, or treatment suggestions, the extra regulatory scrutiny it might appeal to.
That is the place Apple and Samsung’s normal consumer-tech playbook will not be sufficient. Well being AI must be helpful with out being reckless, personalised with out being invasive, and persuasive with out overstating certainty.
The businesses that win this market will probably be those that deal with belief as significantly as options.
Glucose might grow to be the check case for AI well being
AI well being assistants are nonetheless of their early phases, and far of the market stays stuffed with formidable guarantees. Glucose information might power the class to mature.
In contrast to broad wellness scores, glucose readings are particular, steady, and tied to actual well being selections. That makes them precious for AI methods, but additionally more durable to deal with responsibly.
For Apple and Samsung, the chance is evident: construct the AI layer that helps individuals perceive what their physique is doing all through the day. For customers, the query is whether or not these insights are correct, non-public, and clear about their limits.
The smartwatch could not grow to be a blood sugar monitor anytime quickly. However the race to show glucose information into AI well being recommendation is already underway, and the businesses that win it might be those that by no means wanted to measure glucose themselves.
Editor’s word: This text initially appeared on our sister publication, eWeek.
