There’s a particular kind of sinking feeling you get when an app you trust suddenly presents you with an ultimatum. You open Samsung Health one morning — maybe to check your sleep score, log your morning walk, or review last night’s heart rate data — and instead of your dashboard, you’re staring at a consent notice.
.jpg){kind=link}
Not just any consent notice. This one asks you to allow Samsung to use your health data for AI training and modeling. And here’s the kicker: say no, and they’ll delete your data. Not just stop syncing. Delete. Years of step counts, sleep patterns, heart rate variability, medication logs — gone, unless the law says otherwise.
Let’s talk about what Samsung is doing here, why it matters, and why this moment says a lot about where the health-tech industry is headed.
What Samsung’s Notice Actually Says
Over the past few days, Samsung Health users started seeing a notice titled “Consent to the Use of Health Data for AI Training and Modelling.” It’s not buried in fine print — it’s front and center when you open the app. The notice spells out that Samsung will collect health and wellness data, medication data, health records, and cycle tracking data to train its AI models. The company explicitly says this includes “human review” — meaning actual people might look at your health data to improve the algorithms.
There’s a toggle in Settings > Privacy. Flip it off, and Samsung warns you: you won’t be able to sync health data with your Samsung account, and your data will be deleted unless retention is required by law.
No middle ground. No “sync but don’t use for AI.” No “use anonymized data only.” It’s consent or delete.
Why This Isn’t Really Consent
Let me be clear about something: this is not consent in any meaningful sense of the word.
Real consent requires a genuine alternative. Under the Philippines’ Data Privacy Act and every meaningful data protection framework globally — GDPR, CPRA, you name it — consent must be freely given, specific, informed, and unambiguous. When the alternative to saying yes is losing data you’ve accumulated over years, that’s not a free choice. That’s coercion dressed up as a privacy setting.
Think about what Samsung Health tracks for a typical user: sleep patterns, resting heart rate, blood oxygen, stress levels, menstrual cycles, medication schedules, body composition, exercise routes. This is arguably the most intimate dataset a tech company can collect about you. It reveals when you’re sick, stressed, or at your physical best. It knows your habits, your routines, your body’s most private signals.
And Samsung is saying: let our AI train on all of that, or we wipe the slate clean.
What’s Driving This Move
The timing isn’t accidental. In June 2026, Samsung announced a major overhaul of Samsung Health — new features like Vitals (analyzing five health metrics during sleep), Heart Health Score, Daily Cardio Load, and a Fitness Index. These aren’t simple tracking tools. They’re AI-powered insights that need training data. Lots of it.
Hon Pak, Samsung’s SVP and Head of the Digital Health Team, described the vision clearly: “Samsung Health is evolving to connect health data measured by the Galaxy Watch with AI-based insights, enabling users to understand their physical condition more easily and intuitively.”
This was always going to require more data. The question was how Samsung would get it.
With the Galaxy Unpacked event scheduled for July 22 — where the Galaxy Watch 9 series is expected — Samsung is clearly racing to build its AI health models before the next generation of hardware ships. The consent toggle is the data collection mechanism that powers this entire AI health strategy.
This is reminiscent of the broader industry trend I covered in my piece on why companies are done renting their AI — the appetite for proprietary AI models that give vendors a competitive edge is enormous. But when the fuel for those models is your most personal data, the equation changes.
How Others Handle Health Data for AI
Samsung’s approach stands out — and not in a good way.
Apple Health processes the vast majority of its health AI features on-device. Apple’s privacy page emphasizes that health data is encrypted and that the company doesn’t have access to it. When Apple does use health data for model training, it’s typically with differential privacy and explicit opt-in that doesn’t threaten to delete your existing data.
Google Fit and Fitbit (now under Google) have their own privacy challenges, but they don’t present users with a “let us train AI on your data or lose everything” ultimatum. Google’s approach has been more about aggregating anonymized data through its Research Platform, with separate consent flows that don’t threaten your existing data.
Garmin takes a different path entirely. The company has built its reputation on premium hardware and a subscription model — users pay for the service, so Garmin doesn’t need to monetize health data the same way. Their privacy policy explicitly states they don’t sell personal information and that health data collected is primarily used to deliver the service you’re paying for.
Even Whoop, which lives and breathes data-driven insights, has managed to build its AI coaching features without resorting to data ultimatums. Their approach relies on opt-in research programs where users can contribute data with clear understanding of how it’s used.
What Samsung is doing stands out — and not in a flattering way. No pretense of choice. No gradual opt-in campaign. No meaningful middle ground. Just a toggle and a warning.
This matters because health data isn’t like search history or shopping preferences. It’s protected under specific laws in most countries. In the Philippines, the Data Privacy Act classifies health information as sensitive personal information, requiring stricter consent standards. Samsung’s coercive model would be hard to square with that framework if it were challenged. The question of who owns and controls the data that trains AI models isn’t academic — it’s at the heart of the broader discussion about when and how AI will actually pay for itself, and whether users or corporations reap the benefits.
The Deeper Problem: Your Data Is the Product
Here’s what gets me about this — and this is where I shift from analyst to someone who actually manages IT systems for a living.
In my day job at a government ICT division, I spend a lot of time thinking about data governance. We’re implementing data privacy frameworks, securing citizen information, and figuring out how to balance service delivery with protection. Every decision we make starts with the question: “Does the user have a real choice here?”
Samsung’s consent toggle fails that question completely.
The data you feed Samsung Health isn’t just numbers on a screen. It’s a record of your life. Your heart has beaten thousands of times while wearing that Galaxy Watch. Your sleep has been tracked through good nights and bad. Your medications, your cycles, your stress levels — these are not abstract data points. They’re you.
And Samsung is saying: give us all of this to train our AI, or lose the record of it entirely.
This is the same dynamic I explored when writing about trust in AI-driven systems and the security implications of how AI tools interact with our data. When companies hold your data hostage to power their AI ambitions, the trust relationship breaks. And once trust breaks, it’s very hard to rebuild.
Practical Steps for Samsung Health Users
If you’re a Samsung Health user, here’s what you should know:
- Check your privacy settings now. Go to Samsung Health settings > Privacy. You’ll see the “Consent to the use of health data for AI training and modeling” toggle. Take a screenshot of your current state.
- Export your data before making any decision. Samsung Health allows you to download your data. Do this first — it gives you a backup regardless of what you choose.
- Understand the implications of opting out. If you turn off the toggle, Samsung says it will delete your synced health data. Your on-device data may survive, but cloud sync and account-level features will stop working.
- Consider alternatives. If health data privacy is your priority, apps like Health Connect (Android’s centralized health data platform) give you more control over which apps access what. Garmin and Polar also have strong privacy stances, though they come with their own trade-offs.
- Watch the July 22 Galaxy Unpacked event. Samsung’s Galaxy Watch 9 launch will likely clarify how deep the AI integration goes and whether there’s any room for user choice in how data is used.
Where This Leaves Us
Samsung’s move is a glimpse into where the wearable industry is heading. AI health features require training data. The companies that build those features will be tempted to cut corners on consent because the alternative — actually asking users properly and respecting their choices — is slower and more expensive.
But here’s the thing: if every health tracking company adopts the “consent or delete” model, users lose. Not just their data, but their trust in the entire category. And trust is the one thing health tech can’t afford to lose. If people stop believing that their health data is safe, they stop tracking. They stop wearing the watch. They stop engaging with the features that could genuinely improve their wellbeing.
That’s a loss for everyone — users and companies alike.
Samsung has a chance to do better. Remove the coercion from the consent flow. Offer genuine choices: on-device processing, anonymized contribution, or full opt-out without data deletion. Be transparent about what human review actually means and who has access.
Until then, that toggle in Samsung Health settings isn’t really about consent at all. It’s a price tag on your privacy. And the price keeps going up.