Apple Watch Claims 90% of AI Smartwatch Shipments as Rivals Close In

Apple Watch accounted for about 90% of all Edge AI-capable smartwatch shipments in the first quarter of 2026, according to Counterpoint Research. The category grew 70% year over year, and one in four smartwatches shipped worldwide now processes health and safety alerts on its own chip instead of a phone or the cloud.

That dominance rests on a three-year chip lead that is about to face its first real competition. Qualcomm’s new wearable processor is already landing in Samsung, Google and Motorola devices, just months after Counterpoint took this snapshot.

Apple’s Watch Claims Nine Of Every Ten AI Shipments

Counterpoint Research’s Global Smartwatch Shipments Tracker found that Edge AI-capable smartwatches reached 25% market penetration in the first quarter of 2026. Apple took almost all of the growth behind that number.

The firm’s own language leaves little room for debate. Counterpoint said Apple solely accounted for roughly 90% of shipments in the Edge AI segment during the quarter.

Anshika Jain, principal analyst at Counterpoint Research, said brands have been “continuously upgrading their smartwatch hardware to make devices more AI-capable.” She added that Apple “captured the highest shipment share and was the strongest performer in the quarter, helped by its refreshed lineup, meaningful health upgrades and the more affordable Apple Watch SE 3.”

The health features behind that growth expanded fast between Q1 2025 and Q1 2026:

Blood pressure monitoring and sleep apnea detection posted the sharpest jumps. That growth points to health alerts as the reason people are buying these watches in the first place.

What Actually Counts As An Edge AI Smartwatch?

Counterpoint classifies a smartwatch as Edge AI-capable only when it carries a dedicated neural engine or NPU and uses that chip to run at least one health, safety, or interaction feature’s inference directly on the device, not merely include the hardware. That bar excludes plenty of smartwatches most shoppers would already call smart.

Not everyone buys the distinction. Commenters on Apple-focused forums questioned whether Edge AI is a meaningful label at all, arguing that few day-to-day watch features actually run their inference on the device rather than a paired phone.

Analysts covering the report have been careful about the same point. The newsletter Let’s Data Science, which reviewed the tracker’s findings, cautioned that the claim should be treated as a market-share estimate rather than a device-level technical benchmark.

Health Alerts Are Selling These Watches

Health and fitness monitoring remain the single biggest reason people buy these devices, per Counterpoint. Apple’s push here is not new marketing dressed up as innovation.

The company won FDA clearance for sleep apnea detection on the Apple Watch Series 10 in late 2024, well before the feature started showing up broadly in Counterpoint’s shipment data a year later.

Counterpoint’s research director frames the shift as bigger than silicon alone.

Edge AI in smartwatches is shifting from primarily a hardware integration to one that also includes software optimization. The real unlock is smaller, more efficient models and OS-level access that lets any app run inference locally.

Mohit Agrawal, research director at Counterpoint Research, made the point in the firm’s July report on the quarter.

A Three-Year Silicon Head Start

Apple didn’t build this lead overnight. Its Neural Engine architecture first appeared in the A11 Bionic chip in 2017, nearly a decade before Edge AI became a defined smartwatch category.

Apple brought that architecture to the wrist properly in 2023, when the Watch Series 9 debuted with a quad-core Neural Engine twice as fast at machine learning tasks as its predecessor.

Jeff Williams, Apple’s chief operating officer at the time, said the lineup delivered “incredible new capabilities and technological advancements, including a new double tap gesture, brighter display, on-device Siri.”

Huawei didn’t answer with comparable silicon until 2025, when its Kirin W80 chip arrived to run the Celia voice assistant on-device.

Qualcomm’s Chip Lands In Samsung, Google And Motorola Watches

Apple’s run without a direct silicon rival is ending. Qualcomm unveiled the Snapdragon Wear Elite at MWC 2026 in Barcelona in March, its first wearable platform built around a dedicated neural processing unit.

Samsung has already confirmed its next Galaxy Watch will run the chip instead of its own Exynos processor, with launch expected around July 2026. Google and Motorola are also listed as launch partners, though neither has set a release date.

The chip rollout so far looks like this:

• Apple shipped the four-core Neural Engine inside its S9 chip in 2023, the foundation of its current lead.
• Huawei answered in 2025 with the Kirin W80 chip, built to run its Celia assistant on-device.
• Qualcomm unveiled the Snapdragon Wear Elite at MWC 2026, with Samsung, Google and Motorola signed on as launch partners.
• Google is separately developing its own Tensor-based wearable chip, which has yet to ship.

Qualcomm frames the moment as a reset for the whole category, describing wearables as standalone AI computers with their own on-device intelligence rather than accessories tethered to a phone.

Apple’s AI Share Dwarfs Its Overall Smartwatch Sales

Apple’s edge-AI dominance looks even sharper next to its actual sales. The company’s overall smartwatch shipments grew 21% year over year in Q1 2026, giving it a 23% share of the total global smartwatch market, while the entire market grew just 4%.

That 23% figure sits nowhere near the 90% edge-AI number. The gap shows the edge-AI category, as Counterpoint defines it, is still narrow enough that one company’s chip roadmap can dominate it almost by default.

The stakes go beyond bragging rights. Futurum Group’s forecast projects the wider AI devices market to reach $123.6 billion by 2030, with on-device intelligence and privacy named as the core differentiators.

Counterpoint expects Edge AI penetration to approach 32% of all smartwatch shipments by the end of 2026. Whether Apple still holds anything close to 90% of that larger number will depend on how fast Samsung, Google and Motorola actually ship their Snapdragon Wear Elite watches.

Frequently Asked Questions

What Is An Edge AI Smartwatch, Exactly?

An Edge AI smartwatch carries a dedicated neural engine or NPU and runs at least one health, safety, or interaction feature’s inference directly on that chip rather than a phone or cloud server, per Counterpoint’s classification. A software-only alternative also exists: Ambiq’s Apollo platform runs inference on vector-core silicon using Arm’s Helium extensions instead of dedicated neural hardware, though Counterpoint calls it a niche approach compared with Apple’s dedicated-chip strategy.

Which Apple Watch Models Qualify As Edge AI-Capable?

Only models built around Apple’s S9 chip or newer qualify. Apple introduced the four-core Neural Engine with the Apple Watch Series 9 and Apple Watch Ultra 2 in 2023, and that architecture, not the sensors alone, is what lets features like fall detection and irregular heart rhythm alerts run without a network connection.

Does Apple’s 90% Share Prove Its Health Features Are More Accurate?

No. Shipment share measures units sold, not clinical performance. Analysts who reviewed the report have cautioned that the figures are market-tracker estimates, not proof that any specific health model is clinically validated or technically superior to rivals.

When Will Samsung And Google Watches Match Apple’s AI Chip?

Samsung has confirmed its next Galaxy Watch will run Qualcomm’s Snapdragon Wear Elite chip instead of its own Exynos processor, with launch expected around July 2026. Google and Motorola are also confirmed launch partners, though neither has given a specific ship date yet.

What Makes Qualcomm’s Snapdragon Wear Elite Different?

It is the first wearable chip built around a dedicated Hexagon NPU, made on a 3-nanometer process. Qualcomm says it can run on-device AI models with up to 2 billion parameters, extend battery life by about 30%, and reach a 50% charge in 10 minutes.