Ultrahuman Ring Air review

Indian fitness and nutrition tracking startup, Ultrahuman, has fast-followed its debut smart ring last year with a second generation of the device — which officially launched in June.

The new smart ring whittles down what was a slightly chunky form factor, for its original sleep & fitness-tracker, into a more svelte matte black* band it’s calling the Ring Air. So, er, Apple eat your heart out! Although of course Apple doesn’t make a smart ring. And in fact it’s category veteran Oura’s smart ring which most closely resembles what the Indian team has wrought with the Air.

Overview

Ultrahuman may not be so familiar a name to gadget fans but its differentiating edge vs other fitness wearable makers is a metabolic health tracking program, called Cyborg. This extends its ability to provide fitness insights, including by tracking how your body responds to what you’re eating. It does this by looping in another piece of sensing hardware in the form of an arm-mounted continuous glucose monitor (CGM). This medical grade technology, which was originally developed for people with diabetes or pre-diabetes to monitor their blood sugar, is redeployed in this context as a quantified health wearable — with Ultrahuman’s sales pitch promising to turn 24/7 glucose tracking into “instant health nudges” and “metabolic fitness protocols” to help health-conscious consumers meet their wellness goals.

Ultrahuman’s smart ring is designed to sit alongside that CGM-powered program; either as a standalone sleep & fitness tracker (with the bonus ability to dish some dietary insights, if you log what you’re eating in the companion app, powered by aggregated data from Cyborg users); or as a complement to the fully fledged metabolic health tracking program.

The startup’s promise, in the latter case (i.e. wearing both smart ring and CGM) is you’ll be served “deep correlations deep and predictive insights” on how your lifestyle may be impacting your health and fitness goals. Although — one category caveat to bear in mind (which we’ve discussed in earlier reviews) — is that questions remain over whether people without a chronic medical condition really need to be tracking their glucose levels 24/7. Reducing bodily inflammation by maintaining stable glucose may have wide-ranging health benefits over the longer term, is the theory. In the short term Ultrahuman suggests you can expect better energy levels and cognitive performance if you keep blood sugar stable throughout the day. But it’s fair to say that intelligently interpreting this sort of data in an average consumer context remains a work in progress. So this is very much the quantified health movement at its most experimental. Aka, biohacker territory.

Despite it still being early for CGM startups to prove their worth to a general consumer, Oura appears to have noticed what the plucky competition is up. It recently announced it was partnering with a number of other startups that sell CGM-powered tracking services: Namely January, Supersapiens, and Veri — touting the integration as a way for its smart ring users to gain deeper and more personalized insights to help them sleep and train better. Of course Ultrahuman can boast it got there first; last year, when it unboxed its debut smart ring (the eponymous Ring) and begun integrating two tracks of fitness data. Its head-start on metabolic health tracking traces back to 2021 when it launched Cyborg in beta — hence why it’s able to provide Ring Air users with statistical predictions on whether certain foods are likely to spike their blood sugar based on aggregating data from its CGM program’s early adopters (it claims to have amassed more than 600M data-points to power these predictions). Whereas Oura is just starting in on a similar integration, drawing on other CGM startups’ analysis of their users’ glucose response data.

With so much cutting-edge tracking activity circling this segment of the fitness monitoring space, the competition for quantified health fans’ fingers and arms looks fierce. So of course Ultrahuman isn’t sitting still. This sleeker gen-two smart ring looks intended to close the gap with Oura’s hardware. The Ring Air starts at $349 for the matte black version (which has a Titanium body coated in Tungsten carbide) vs a slightly cheaper starter price-tag ($299) for Oura’s most recent smart ring. However — unlike with the Oura Ring 3 — there’s no monthly “membership” subscription required for ongoing tracking with Ultrahuman’s Air. It’s a one-shot payment for all the Ring Air’s core sleep and fitness tracking features. That simplicity in pricing may appeal to consumers tired at the thought of yet another monthly subscription to shell out for.

If you do also want to fold real-time blood glucose insights into your quantified health mix you’ll need to pay extra to be shipped the necessary sensors (a pack of two CGMs, good for a month’s tracking, costs €189 in the EU). But, as noted above, even if you don’t shell out to monitor your blood glucose via arm-mounted sensor the app can draw on aggregated insights from others — serving up predictions on how you might respond to foods as you log them. (Though you also don’t even technically need to own its smart ring to tap into this feature; the app is free to download.)

If you have previously been a Cyborg user Ultrahuman suggests you’ll get an even richer experience, since it can tap your historic blood glucose response data to power more personalized predictions without you needing to buy fresh sensors. Or, well, that’s the claim. It gives the example of a customer who previously logged a glucose spike after eating a slice of pizza, explaining: “The next time you log a slice of pizza, the app will nudge you to achieve personalised movement goals to manage the spike based on your unique profile and movement data.” But this feature didn’t seem to trigger such specific feedback during our testing. 

In any case, it’s important to note that a person’s own glucose response to eating the same food isn’t necessarily going to be identical. It can vary — based on things like when and how quickly they ate, what they ate earlier and even how much sleep they’ve had, whether they’ve exercised recently, or how stressed they feel — so your own historical logs also aren’t necessarily a perfect guide to whether or not today’s meal will spike your blood sugar. Which is why there can be worth in doing continuous glucose tracking periodically, or at least more than once, to compare your own responses now vs then. 

In short, glucose tracking is complicated — even messy. And figuring out how best to act on the data can be challenging. On the surface Ultrahuman’s smart ring side-steps that complexity, offering a more familiar set of core fitness-tracking features (movement, sleep, recovery). So far, so typical to the category. But by folding everything into the same app it extends the feature-set available to its smart ring users — who can also use the software to log what they’re eating and be served predictive insights on whether their meals are likely to keep their glucose stable. So the really big nudge Ultrahuman looks to be hoping to pull off is to cross-selling gadget fans on the value of taking an even deeper, real-time dive into their biology by slapping on a CGM too. 

Read on for our full review of what the new Ultrahuman Ring Air can do…

*Ultrahuman says other colors will follow the debut matte black Ring Air, starting with a “mirror finish” silver option

Ring vs Ring Air

Old vs new: Ultrahuman Ring (gold) and Ultrahuman Ring Air (black) shown next to each other for comparison (Image credits: Natasha Lomas/TechCrunch)

The first noticeable difference between Ultrahuman’s (debut) Ring and the Ring Air is a more slender form factor that makes the new device easier to wear. The Air is also extremely light (weighing in at just 2.4 grams). So it’s certainly Ultrahuman’s most wearable smart ring yet.

Wearability here means you can rock it on your finger and get on with your day without having to take it off too often because it rarely feels like it’s getting in the way. It’s also minimal-looking enough not to attract comments about your flashy hand bling. (Whereas the chunkier Ring, which I tested earlier this year — albeit, in a gleaming gold color — did get remarked upon.) I also often ended up wearing the first-gen Ring on my thumb because the bigger profile didn’t always play nice with everyday activities like chopping veg or unscrewing a stove pot to make coffee. But I’ve never felt I needed to wear the Air on my thumb.

The Ring Air’s fit is noticeably snugger than the original Ring. So despite both Ring and Ring Air technically being the same ring size the newer band seems slightly smaller as it fits exactly on the index finger where it was intended. At a pinch, I could wear it on my thumb but this time it won’t slip on unless I apply some force to push it over the knuckle so Ultrahuman’s ring sizing seems truer with this second generation device.

One gripe on fit: The design of the Ring Air’s inner band includes a slight convex bump at the bottom where some of the sensors are located — presumably so they press against the finger to ensure close contact and better readings. I found this bump plus the snugger fit translated into a mild pinching sensation which was just the wrong side of comfortable. I’m not saying it’s actively uncomfortable and I did get used to it over time. But nor is it “ultra-comfortable”, as Ultrahuman’s marketing suggests. I did also notice a bit of skin discoloration developing on the fleshy part of my finger, akin to what you’d see under a bandaid after you take it off. This was after wearing the ring for a few days. It didn’t itch or feel painful but the mark was a little disconcerting. It ended up fading away after a while so may have been related to moisture being trapped under the band.

Ring fit and skin issues are obviously personal so your bandwidth for experiencing any pinching or other discomfort will vary. But it is one consideration to bear in mind if you’re weighing up whether to buy a smart ring or a smart watch. For the former, fit may be more challenging in summer given fingers can swell when it’s hot. (And I was testing the Ring Air during some of the hottest weeks of the year — indeed, possibly, the hottest month ever since humans began recording temperature.) With such a snug fit this time I found there were fewer options for switching the Ring Air to a different finger. Ergo choose your ring size carefully. (NB: Ultrahuman sends out a sizing kit of plastic dummies before shipping the ring in your chosen size.)

The next patently obvious difference between the two generations of Ultrahuman’s smart ring is battery performance. Despite a (marginally) larger form factor the first-gen Ring suffered a bit on this front. If you got through a full day’s wear on a single charge that was about as good as it got (I did also have to get a replacement for the first unit after a few weeks after the battery failed entirely). Whereas the Ring Air is able to last days, plural, on a single charge. Up to six days’ use is Ultrahuman’s claim.

I maybe didn’t get quite that much usage out of a single charge but battery life was certainly way better than before. Performance also seemed to keep up over the (several weeks) testing period. Although charging itself feels a little slow. A couple of examples: After charging to full when the ring first arrived on a Wednesday afternoon I didn’t need to recharge it until the following Monday morning. Another time a full charge on a Saturday morning lasted until Thursday morning (when it still had 3% charge at 08:30). Although Ultrahuman recommends not letting the battery get anywhere near that low, suggesting it’s kept between 30% and 80% to help extend battery life. So, in practice, you’ll probably only be able to go max 3-4 days between charges if you’re going to stay in the recommended optimal charging window. (At one point the app did pop up a notice urging me to use shorter charging bursts, couching charging more as topping up, so the actual ideal is probably to do a bit of daily charging, wherever possible.)

As with the Ring, the Air’s finish obviously won’t be immune to your lifestyle. And after about a week’s wear I found the Air had, under closer inspection, acquired a handful of scuffs. So expect some patina to develop. The matte black coating was perhaps a little more resistant to marking than the shiny gold of the gen-one Ring. Although scratches do seem to show up more on glossy gold than matte black. So expect different finishes to have their own quirks.

Fitness & wellness features

The main features Ultrahuman lists for the Ring Air are: 24/7 tracking of sleep, movement, heart rate (HR), heart rate variability (HRV), and skin temperature. As well as Sleep and Movement Indexes, this tracking mix also translates into a (post-sleep) daily Recovery Score to help you plan your training or other physical activity. Basically a lower score here means you might want to take it easier on workouts that day vs a higher score being a signal to try for your personal best at the gym/running track etc. (This feature will be familiar to Fitbit, Oura and Whoop users, among others.)

The sleep and movement indexes are the most self-explanatory main features. Ultrahuman’s sleep tracking detects different sleep stages, such as deep, REM sleep etc — so the morning after you’ve clocked up a decent amount of shut-eye the Sleep Index might greet you with a high score plus an affirmation like: “Optimal sleep volume”. Conversely, if you skimped on hours in bed and/or slept poorly you’ll be nudged to prioritize getting enough quality Zzzs to boost your rest and recovery.

In addition to nightly sleep tracking there’s a nap detection feature for power naps over 25mins. I’m not a fan of naps but took one for the team to test this and found it did correctly detect a nap (note: there is a short lag before it pings you to ask you to confirm/deny a siesta). Albeit, another time, it incorrectly suspect me of napping — although I had been reclining and not moving for a while which presumably confused the algorithms.

Another sleep-related feature is circadian rhythm tracking: The app will nudge you to get exposure to natural light or avoid bright light during certain time windows in the morning and evening to support better sleep quality — displaying a watch-style digital dial to indicate which circadian phase you’re currently in. This would, ironically, probably feel more useful displayed on a smart watch where you can glance at it, rather than in an app on your phone.

As well as nudging users to achieve “sleep schedule consistency” (i.e. getting to bed and waking up around the same time), the other use-case Ultrahuman touts for this taps into the phenomenon whereby exposure to light (and being active) during a “phase advance window” can help you move your sleep-wake cycle earlier in the day — which may be useful if you’re trying to get over jet-lag.

The Movement Index probably needs the least explanation. The Ring Air basically tracks how sedentary or otherwise you are. This means if you’re sitting at a desk typing all day you can expect to get low scores. But if you at least get up and move around a bit every hour, as the app nudges you to (dispensing “stretch break alerts”), you’ll chalk up a slightly less depressing performance. Whether humans really need a pricey gizmo to tell them to get off their backside is one (category) question to ponder.

I still find movement tracking and nudges the least interesting aspect of fitness trackers. Sure, they have to include it as its table stakes for this sort of product and the data obviously feeds wider lifestyle tracking capabilities. But unless you’re a secret sleepwalker the insights aren’t going to surprise you so it’s hard for the Movement Index to feel engaging. And quite often I found it a bit frustrating. Not just because my job requires me to be a desk slave; but for certain types of physical activity, such as climbing or weight lifting, you’re going to need to remove the ring to do your workout (Ultrahuman explicitly warns against wearing the ring to lift weights in case of scratches). Which means the hardware can’t track any of that exercise in real-time. 

If you also have a smart watch you can use for tracking workouts you may be able to opt for data to be shared with Ultrahuman’s app to avoid workout logging blackouts (it can sync data from the Apple Watch, for example, if given relevant permissions). You can also track activity after the fact in Ultrahuman’s app. But that feels a bit more like a chore vs doing it in the moment and getting richer workout data. Of course a smart watch can just remain on your wrist the whole time you’re at the gym, quietly tracking everything — so the smart ring form factor does have some drawbacks. 

Sleep and recovery tracking are, ultimately, where the real action is with Ultrahuman’s smart ring. As noted in our review of the debut Ring, it goes deep on sleep tracking — certainly you get a more nuanced experience than the pretty basic tracking offered by a more mainstream consumer wearable like the Apple Watch. The greater focus on tracking sleep quality also enables Ultrahuman to calculate the aforementioned daily Recovery Score — which essentially quantifies how well rested you are (and, thus, how ready to push hard in your next workout). So while I do continue to struggle, somewhat, with the idea we all need to be wearing a sensing band to bed so an app can inform us how well we slept (when our body will tell us that if only we’ll listen to the signals), the wider utility of sleep tracking is clearer when, as here, it’s integrated into a broader mix of tracked data. Certainly, the insights being surfaced do start to feel richer.

That said, how useful the Recovery Score feature will be for you really depends on your lifestyle. This sort of tracking is most obviously geared towards more committed athletes who are looking to push their edge and/or training for a particular goal/competition. Although there are quite a lot of products catering to pro fitness folks, and a smart ring form factor isn’t the immediately obvious choice for the segment — so winning over the most athletic users seems likely to be a challenge.

For a more general consumer who just wants to improve their fitness level and generally be more active the depth of tracking may feel a bit like overkill — i.e. if you’re just trying to carve out half an hour to get to the park for a jog, rather than training to smash your personal best.

Add to that, if your lifestyle isn’t conducive to regularly getting good rest and recovery then being systematically slapped with poor scores in the Sleep and Recovery Indexes every morning isn’t going to feel great. It may even feel demotivating, potentially generating more stress than it’s able to save you.

The ideal Ring Air user is probably someone who’s already pretty keen to sustain and/or improve their fitness level. To get the most out of the product you do for sure need to be motivated enough to put some effort in to get your metrics moving in the right direction. Which means being willing to make some lifestyle changes, such as picking healthier food choices. Or at least moving when nudged and making tweaks to stuff like when you go to bed. Committed loungers, candle-end burners or party animals will probably just find the whole experience a terrible nag/buzz kill.

Another quick note on form factor: With the smart ring being so (relatively) small and light you can easily forget to put it back on if you’ve had to take it off (say to wash or for exercise). And if you do that and end up forgetting to wear it overnight you’ll obviously wake up to no Sleep or Recovery scores. Which happened to me at least once. Doh! This kind of self-imposed tracking blackout is an easier pitfall with a smart ring than a smart watch IMO. Unless you routinely need to charge your smart watch overnight. If that’s the case the Ring Air’s longer battery life is a convenience boon. Plus the smaller form factor may be more comfortable to sleep in vs a more bulky smart watch.

Image credits: Natasha Lomas/TechCrunch

Where I’ve found Ultrahuman’s smart ring tracking most useful is in understanding how different lifestyle factors can negatively impact sleep and recovery. And, by implication, get in the way of wider fitness and health/wellness goals.

After wearing the Ring Air for a while you should start to find yourself noticing patterns — such as low scores in the sleep and recovery indexes after eating a late meal, say, or doing a late workout. Or indeed working late. Drinking alcohol and experiencing a stressful event can also depress these scores. On the flip side, you may notice you’re seeing much better scores for sleep and recovery after eating dinner earlier than usual, or picking healthier food and drinks choices, or just making the effort to get an early night — following the app’s pings to “start winding down” and avoid screentime before bed.

The key thing here is you do need to make an effort to engage with the tracking. At least by checking in on your Sleep and Recovery scores in the morning. Happily Ultrahuman’s product design makes that bit easy. So it’s pretty simple to start joining the dots of what you’ve been up to recently and how pleased (or otherwise) your body is about it. So you’ll quickly find yourself building up a bank of personalized insights that illustrate/reinforce how various lifestyle factors can have a noticeable impact on how well you feel (and, indeed, possibly a lot more than that).

The best thing about this is it really doesn’t require a lot of effort because core components of the tracking are passive. All you need to do is check in with the app now and again and make the mental effort to interpret the daily scores in the context of what you were up to — after which you can just get on with your day. Perhaps now fired up with the data-driven motivation you needed to make lifestyle adjustments to see if you can edge towards better scores tomorrow or the next day.

Ultrahuman’s app surfaces an at-a-glance summary of your Sleep and Recovery indexes every morning (at least assuming you didn’t forget to wear the smart ring to bed!). This provides a score (out of 100) plus some key highlights and takeaways. For example, you may be greeted with a top-line note saying: “Optimal deep sleep for physical recovery”, plus a brief explainer: “Higher deep sleep helps with physical recovery and immune system health”. Which might be more than enough for you to get on with your day.

But if you do want more detail you can drill down into the index to view ratings for individual factors contributing to the overall score — which can help troubleshoot why you might have got a poor score. So there are lots more data points to be got at if you do want to geek out further. Even as the average consumer can be reassured the app does the legwork of crunching your data to generate daily reports to help you understand if you’re on track.

One caveat that generally applies to consumer fitness trackers, whether smart rings or smart watches, is around the accuracy of the data. Bottom line: These are consumer products (and use-cases), not medical devices or services. So we can’t know how on point (or not) any of these measures and metrics really are. Although it’s an area where rivalry in the wearable category looks set to heat up.

Ultrahuman’s smart ring hardware is no different in this respect. (Unlike the medical grade CGMs it repurposes for its metabolic tracking program — which are made by US medtech giant, Abbott.) And it explicitly warns that issues like poor fit can affect the accuracy of readings. It also suggests a two-week usage “buffer” to give the ring’s sensors time to gather enough data to calibrate readings, ensure “richness” and “increase accuracy”, as it puts it. And while it does specify that the Ring Air’s sleep-tracking functionality has been benchmarked with EEG Devices — “ensuring reliable and accurate monitoring of sleep patterns”, per its marketing claims — if the sensors have poor contact with your skin because the device is a bit loose on the finger you wear it on it’s not going to be able to log the best quality data. So unknowns on accuracy are a given.

At the end of the day, it’s simply not possible to robustly verify the output of the tracking, short of taking the ring into a lab and performing comparative tests with medical grade trackers. Which we have not done.

Anecdotally, I will say Ultrahuman’s scoring of my sleep and recovery did seem greatly improved vs when I was testing the beta release of the gen-one Ring. With that device I was baffled by how often my sleep and HRV scores, especially, did not match how well rested and recovered/energetic I felt. But this time the Air’s scoring seemed much closer to my reality. And my gut feeling is the tracking has gotten a lot more accurate. (One anecdote: After waking from a sleep filled with particularly vivid dreams the app informed me it had detected a lot of REM sleep which seemed eerily spot-on.)

Ultrahuman did eventually confirm there had been a bug with the gen-one Ring which had caused me to receive consistently low HRV scores — and pushed out a fix. But how much you can trust the data that consumer-grade fitness trackers are generating remains a category question mark.

As a general rule, these devices are best treated as directional fitness guides — to whether you’re on the right track or should be taking a closer/more critical look at your lifestyle. You absolutely shouldn’t treat the metrics and measures they serve as ‘gospel’ quantified health truth. At best it’s signposting and the products are tools which, if used carefully, may help you move closer to your goals.

Almost all of the features I’ve discussed so far were familiar to me from Ultrahuman’s debut smart ring. But it’s worth noting that pushing out regular software updates to build out the hardware’s capabilities is a staple of how Ultrahuman operates (you can find its roadmap for the smart ring here). That means any review is necessarily just a snapshot of a moveable feast. So you should expect continued evolution, including tweaks to how data is presented in the app.

On the latter front, the current iteration of the app includes more detailed and approachable explanations of different data-points and why it’s useful to measure them (and/or how you may want to respond to your results) vs earlier versions. So the product dev team looks to have focused on boosting usability by helping users better understand how to decode their data. Which is very welcome. Although more basic accessibility doesn’t seem to have had as much attention since a lot of these (often) dense text explainers use small grey fonts that can be hard to read on a smartphone screen.

In-app views: Ultrahuman’s circadian rhythm tracker, Sleep, Recovery and Movement Indexes (Image credits: Natasha Lomas/TechCrunch)

So what’s new to the Ring Air in terms of features? One new feature that launched while I was testing the device is a workout mode. This is different to the activity logging you could already do as it’s intended to capture workout metrics in real-time. Ultrahuman says it provides “precise heart rate tracking during active workouts”.

The workout feature looks intended to compete more directly with fitness tracking smart watches like the Apple Watch by providing users with an in-the-moment view of how their body is coping with exercise (HR, cardiovascular response, workout intensity levels etc); and the ability to check in on and track these metrics after the fact — with its pitch suggesting the feature supports decision-making to “optimize” fitness outcomes. 

That’s the theory. Thing is, a smart ring isn’t a smart watch. It has no screen so can’t be glanced at to get a quick eyeful of real-time data in the way a well-designed smart watch can. So it’s not clear how immediately useful a workout tracking feature is in this context. At least, not in terms of helping you keep tabs on exercise performance in the moment since there’s no easy way for a Ring Air user to monitor the data as they sweat — short of having the app open on your phone and your phone in your hand or at least in eyeshot. Which just sounds annoying for a lot of exercise scenarios.

Given that, I question how useful this feature will be during workouts. Of course the structured data will surely provide Ultrahuman with a wealth of extra intel on bodily responses to exercise which could support ongoing feature development (and boost utility in the future). But for users actually doing a workout the feature will probably feel a bit underwhelming. And won’t seem like a direct replacement for the kind of real-time workout tracking you get on a well-designed, fitness-focused smart watch. It’s just obviously more useful to have key metrics like HR and workout zone displayed on your wrist while you’re exercising in most activity scenarios.

Of course the existence of a workout mode does at least mean Ring Air users can directly track their exercise metrics — which shrinks the utility gap with activity-logging smart watches. Although, to be clear, you’re not getting a whole separate fitness app for workout tracking; Ultrahuman folds workout logging into the general timeline. After you’ve logged a workout, it’ll be recorded in the app’s event feed where you can tap on it to drill into the data. The app displays a breakdown of workout zones; max HR; average HR and HR over time; total calories burned and over time; as well as charting heart rate recovery for each exercise session logged. (To log a workout you hit the plus (‘+’) icon above the event timeline which brings up a menu with the workout tracking option. This is also where you can log meals, activity or input other lifestyle-related data points.)

I tested the Air’s workout feature directly against the Apple Watch to see how the devices’ exercise tracking compared. Perhaps unsurprisingly, there were marked differences in the data each recorded for the same workout. For example, when it came to workout zones, the Air logged more minutes (22:14) in zone 1 (<130BPM) for the same HIIT workout as the Apple Watch — which only recorded 15:40 in zone 1 but clocked more minutes (09:20) in zone 2 (131-141BPM) compared with just 04:02 in zone 2 recorded by the Air. Apple’s smart watch also recorded slightly more time in zone 3 (142-153BPM) than the Air. But Ultrahuman’s smart ring clocked a tiny amount of time in zone 4 (154-164BPM) vs zero minutes recorded in that zone by the Apple Watch.

A yoga workout also threw up some oddities — with the Air recording a notably higher average HR (97BPM) than the Apple Watch (87BPM) for the session. Presumably because it registered two brief periods of heart rate elevation that probably didn’t occur (recording 30 and 40 seconds respectively in zone 3 and zone 4 workout zones, whereas the Apple Watch recorded the entire session as a zone 1 workout). Yoga is not typically an intense workout so I’d expect it to be zone 1 only, which suggests the Apple Watch’s reading was the truer one (and the Air likely made a couple of rogue recordings).

The Air also recorded similar (brief) elevated heart readings for an outdoor walk I logged, whereas — once again — the Apple Watch tracked it as only zone 1 exercise. So the discrepancy in workout tracking wasn’t a one off. And — to my eye — Apple’s data looks more solid. Also because during some workouts, I noticed that the Air’s heart rate tracking could fluctuate quite a bit — sometimes dropping from an elevated rate down to something closer to resting heart rate in the middle of intense exercise, seemingly for no reason.

I suspect Ultrahuman has more work to do calibrating workout tracking. (It does warn that: “The readings obtained in the first couple of weeks are required for calibration and hence, it is not recommended to take those values at face value.”) But such discrepancies do just re-emphasize the accuracy point made above. And almost all the readings you get from consumer fitness trackers should be treated as unproven. 

Another new feature that launched during testing focuses on (legal) stimulants (e.g. tea, coffee) and when to take them. Ultrahuman calls this a “stimulant window recommendations” and says the idea is to help users make the most of the productivity boosts caffeinated drinks can given you (including by using them as fuel for workouts), as well as avoiding poor intake timing meaning your coffee habit interferes with sleep/recovery etc.

What this boils down to is some pretty rudimentary in-app signalling/nudging to avoid drinking coffee (or similar) too soon after you wake up in the morning (“to let residual adenosine clear, and heighten the stimulant effect during your productivity peak”, per Ultrahuman’s explainer); plus encouragement to “taper off” your cup(s) of joe before it’s too late in the afternoon to avoid interfering with a good night’s sleep.

As someone definitely on the sensitive-to-caffeine side I was curious to see what revelations the feature might serve up. But it didn’t really tell me anything I didn’t already know. Or at least wasn’t already doing myself (since I don’t drink coffee/tea etc first thing nor late in the afternoon anyway). So aside from an initial boost of reassurance that my approach to caffeinated beverages is a solid one, it didn’t offer anything in the way of ongoing actionable value. The upshot was the feature feels kind of gimmicky; like a bit of (standard) advice has been dressed up with a few in-app bells and whistles.

More new features are on the way to the Air, per Ultrahuman. And — potentially — some more substantial features. Two additions slated as incoming soon (this month) are a measure of cardiovascular fitness (VO2 max) and real-time respiratory rate tracking.

VO2 max has become a sort of vanity metric in the world of fitness trackers. It’s supposed to be a measure of the maximum amount of oxygen your body can utilize during intense exercise. Although consumer grade devices are only estimating this, based on tracking signals like HR and motion data, rather than performing an exact measurement of your body’s oxygen uptake efficiency. So, as ever, what they’re offering here is just general signposting on whether your fitness is improving (or falling), rather than a status to be printed out and stuck up on the wall as a sign you’re pro athlete material.

At the time of writing Ultrahuman’s VO2 max feature hadn’t yet launched. So it will be interesting to see how it compares to the Apple Watch’s version of the feature, which it labels as a measure of “Cardio fitness”.

Ultrahuman says its forthcoming VO2 feature will help Air users improve their cardiovascular fitness and overall health by providing “personalized recommendations tailored to your fitness goals”. And in what looks like a twist on what Apple tracks, it claims it will also be able to assess arterial stiffness and aging — saying the former “indicates generalized thickening of your arterial wall, which may give insight into high blood pressure or hypertension”.

To be clear, it’s not saying the Ring Air will be able to measure your blood pressure. But the suggestion is it might be able to offer an indication of possible blood pressure issues. (“Inflammatory arterial change can lead to dysfunction of the endothelial walls (a single-cell layer that lines all blood vessels and regulates exchanges between the bloodstream and the surrounding tissues) and excessive deposition of lipids or fats in the walls,” Ultrahuman also offers to explain why tracking this might be useful.)

As for the real-time respiratory rate tracking feature, which also hasn’t yet launched, Ultrahuman suggests it will help Ring Air users better understand their body’s response during exercise, as well as quantifying their “breathing efficiency” — suggesting it can be used to “optimise breathing techniques, enhance endurance, and improve overall respiratory health”.

Tracking breathing during exercise might be especially useful for runners or others focused on cardio sports. Although plenty of smart watches can monitor respiratory rate and at least provide an average so you can keep tabs on any changes. So how much extra utility Ultrahuman’s “real-time” tracking might offer remains to be seen.

Bonus food insights

Another aspect of Ultrahuman’s app that Ring Air users can tap into is focused on food. The app provides tools for you to log what you’re eating — displaying a (basic) running total breakdown of your daily macros if you start logging your meals. (But of course this will only be useful if you’re being precise with your food logging by inputting both ingredients and quantities/weights consumed.)

Macro tracking aside — which won’t be for everyone, obviously — there is a reason to at least log what you’re eating in Ultrahuman’s app. Because, as noted above, it bakes in data-driven predictions than give you feedback on whether your food choices are more or less likely to spike your blood sugar, drawing on insights its gleaned from its Cyborg program.

This novel feature offers a taster of how different foods may impact glucose regulation — giving you some pointers on stuff to avoid (or eat less of) and foods that might be more likely to help reduce inflammation and be a good addition to a healthy diet.

That said, the advice can be very general and doesn’t always feel that individually useful. So, sure it might be interesting (if unsurprising) to know that 77% of Cyborg users got a stable glucose response to consuming broccoli — so probably you will too. Or that 64% had a stable response to eating a poached egg. Yay! But it feels a bit less useful to know that 54% of users got a poor response to coffee with milk. Hmm. Or even that 50% had a stable response to pears — which suggests the other half didn’t. But, well, I guess you could flip a coin to decide whether or not to eat a pear. Or just eat half a pear instead of a whole pear?

As noted earlier, many factors impact glucose regulation, not just the food you’re eating. So it’s a complex type of prediction to pull off, even with hundreds of millions of data-points to feed off as Ultrahuman is here. Bottom line: Your glucose response to specific foods won’t necessarily match these predictions.

I did find the predictions I was served didn’t always match up with my (actual) responses (as registered by CGM). Including in the case of a couple of slices of fresh fruit (watermelon) that had been predicted as probably bad for blood sugar stability — because 67% of users did not have a stable response to eating it. But in my case eating watermelon seemed fine. So go figure! (Or I just ate less watermelon than others had at a sitting?)

There’s a further wrinkle if you don’t take the time to individually log every single food item/ingredient you consume — and I certainly didn’t always do that because you can save time by logging your meal as a single entry with multiple ingredients — you’ll be served with an even fuzzier prediction, such as “67% of users got a poor glucose response from one or more items”. This doesn’t tell you which of the items the prediction refers to. Which is obviously even less enlightening.

Ultimately, the only way to get a true view of your body’s glucose responses is to strap on a CGM (or get comfortable with finger pricking and a blood glucose meter). So maybe the biggest value in Ultrahuman having baked these predictive glimpses into its app is nudging a wider pool of users towards taking that semi-invasive step. So it looks like a smart upsell for Cyborg, as well as a free value-add for its smart ring users.

That said, there are some fairly glaring bugs in the food predictions interface. For example, it will sometimes generate a predictive score presumably for the food you’ve just logged but display the name of a foodstuff you logged earlier. Argh! So you have to drill into the interface to try and get a clear read-out. (That seems to especially be an issue when you’re logging a number of items in fairly close order.)

Beyond that, the food logging interface generally remains the most finicky and the least user-friendly aspect of the app. And more work is needed to polish this side of the product experience — which is actually an essential component if you are signing up as a Cyborg (see below).

I also want to quickly mention another food-related feature that was new to me in the app is called an “AI food optimizer”. This uses generative AI (apparently based on ChatGPT) to offer up some ideas for how to optimize your meals to lower the risk of glucose spikes. So, for example, if you’ve logged a veggie stir fry with rice it might suggest switching the rice for an alternative with a lower glycemic index and adding some lean protein to help slow the rate of digest and reduce the glucose spike.

Unlike the (human) in-app coaches Ultrahuman (also) makes available for users to ping for advice, this generative AI tool boasts absolutely no qualifications in nutrition, sports science or anything else. So you should obviously take the AI’s “advice” with a big pinch of salt. (And a bunch of the stuff this serves is very generic in any case.)

At least Ultrahuman has added a warning (albeit in very small print) about inaccuracy/unreliability which appears at the bottom of the automated recommendations. But whether something as important for health as dietary guidance should be left to a mindless AI is one question to consider. And it may be advised to dial up the prominence of the disclaimer.

Ring Air + CGM

Tracking your blood glucose via CGM and Ultrahuman’s app is a whole (other) product experience in and of itself! We’ve covered its “Cyborg” metabolic tracking program before, when it was still in beta here; and again when Ultrahuman launched its debut Ring — and I remain convinced the company is onto something of broad value here — but for this section we’re just going to focus on what’s new with the integration between smart ring and CGM sensor. (If you’re looking for an overview of what to expect from tracking your glucose with Ultrahuman’s CGM-based program we encourage you to read our prior review.)

The first thing to say is the link between the two products remains fairly subtle from the user’s perspective. It won’t be immediately obvious how the respective data streams are being integrated (aside from a widget in the Air tab where the app displays a brief summary of your live glucose and metabolic score, and — if you’re doing well keeping your blood sugar on the level — you might also see a note about your relatively performance vs other Cyborgs, such as that your score is in the “top 20%”). But, frankly, you’ll probably be too consumed by the app’s live glucose tracking to notice. Watching how your blood sugar reacts to different meals and activities can be pretty fascinating/horrifying.

This time testing the product, I actually felt as if the integration between smart ring and CGM had been dialled back compared to when I was testing the gen-one Ring as I seemed to be getting fewer notifications based on cross-linking the two types of wearables’ inputs. For example, I didn’t see any pop-ups flagging a “good spike” this time around — i.e. linking a rise in blood glucose to an exercise session — despite the app (now) being able to log your workouts in real-time.

When my blood glucose spiked during exercise Ultrahuman’s app just noted a “hyperglycemic event” in the timeline but didn’t offer deeper contextualization, despite also having the data inputs to know I was engaged in some intense activity at that time. Which seems like a step backwards to a less integrated product experience than before.

I also didn’t spot any richer nudges connecting being more active to better glucose outcomes — such as pings I’ve had from product previously which affirmed that an after dinner walk I took had helped managed the size of the glucose spike.

I don’t know if this is because my glucose responses have actually been more stable vs when I was last testing the product combo. It’s possible there have been fewer opportunities to be served those kind of nudges this time. But my sense is the product dev team’s attention has been focused elsewhere — on finessing the UX for the Ring Air — so the more complex and challenging work of intelligently interpreting and integrating signals from the different devices into a coherent user experience seems to have been parked.

The app does have a bunch of my historic glucose response data but also wasn’t obviously joining the dots between that to what I was logging eating this time — i.e. to offer guidance on how active I should be to help maintain stable glucose, as Ultrahuman claims the Ring Air + CGM wearable combo can. The activity nudges it did serve seemed a lot more rote and generic than that — such as reminders to “stretch you legs for a quick energy boost” every few hours or so.

So while Ultrahuman’s marketing claims that wearing both its smart ring and CGM will unlock “deeper correlations” and more personalized predictions that still feels more like an ambition at this stage. To my eye the integration between the two products remains a work in progress and the UX here is immature.

As noted above, food logging is also still the most wrinkle-prone/buggy element of the app. And if you’re wearing a CGM you will have to use it so there’s no avoiding the most frustrating bits of the product experience. 

Another bug that bugged me was the app has a mind of its own about how to group food logs for scoring meals, especially if you’re not spacing out your meals/snacks that widely. Frequently, it was bundling lunch, snacks, drinks and even dinner into one long all-day mega stack — which maybe got scored out at the end of the day or sometimes the app just wigged out. But a score on everything you ate across the whole day isn’t really useful. 

The app has always had a tendency to do this over-bundling. But, if anything, it seems to have become more of an issue than before. So, again, this aspect of the product experience doesn’t seem to have had as much love as the smart ring tab clearly has. Hence my sense that the CGM side has been left to largely tick over while Ultrahuman goes all in on raising its smart ring game. 

Bottom line

Ultrahuman has delivered a significant upgrade with its second generation smart ring. There are major improvements to the hardware and software, including a slicker user interface that does a much better job of signposting key takeaways and serving feedback that’s easier to digest and act on. Core fitness tracking features, such as sleep and recovery, also appear to perform more solidly than with the gen-one Ring. So, overall, Ultrahuman’s hardware feels ready to stand up and be counted in a category that contains a polished veteran like Oura.

At the same time, certain newer features don’t seem as solid, nor as useful or reliable as the core set yet. Plus there’s still a fair amount of experimentation on show. And we do wonder if the startup is spreading itself a bit thin. Ultrahuman will need to keep honing and proving out what it’s building to ensure it can deliver robust utility — avoiding the temptation to veer into gimmicks in a bid to stand out from the fitness tracking competition. It’s also hard to avoid the (category-wide) conclusion that for many consumers a smart watch will simply remain the more practical choice for fitness and workout tracking needs.

When it comes to the more experimental (and still niche) arena of glucose tracking, Ultrahuman continues to work a differentiating edge. But it’s now dealing with an obvious case of split attention as it works to ensure its smart ring packs a punch. So while its app does a solid enough job of presenting glucose data and providing useful functionality like rating meals and scoring how well fuelled (or not) you were during a workout, there are some annoying interface bugs. It’s also less clear how — or even if! — it’s joining the dots between the smart ring’s tracking and CGM data. So the promise in its marketing of deeper insights by wearing both doesn’t yet stack up.

Decoding glucose response data to tease out actionable insights remains Ultrahuman’s stated mission — to combat what it has described as a global “metabolic health crisis”. But its path toward that ambitious goal has forked into building out a host of (more mainstream) fitness tracking features first. So we predict that delivering on its full-spectrum metabolic health vision is going to be a marathon, not a sprint.