Apple Watch Series 12: The Future Arrived, Then Tripped Over Its Own Feet

Apple dropped the Series 12 yesterday (September 9, for those keeping score), and I've spent the last 18 hours reading every spec sheet, watching every hands-on video, and arguing with strangers on Reddit about whether ceramic is worth the upcharge. Short answer: maybe. Long answer: Apple's given us some genuinely clever improvements wrapped around some utterly baffling omissions.

Let's start with what's actually interesting.

The S11 Chip and On-Device AI (Finally)

Apple's been dancing around proper on-device AI for years now. Siri's been rubbish since launch, requiring an internet connection for the simplest tasks while your phone sits there with a neural engine doing nothing. The Series 12 changes that. The new S11 chip handles Siri requests locally, processes voice commands faster, and (according to Apple) learns your health patterns without sending data to the cloud.

This matters more than you'd think. I've been wearing smartwatches since the Pebble days, and the single most annoying thing about every Apple Watch has been waiting three seconds for Siri to ping a server just to set a timer. My Garmin does it instantly. My £200 Amazfit does it instantly. Now, finally, the £399-and-up Apple Watch will too.

The AI angle extends to health tracking. Apple's calling it a "Redesigned Health Sensing System", which is marketing speak for "we sample your heart rate more often and crunch the numbers locally". You get a Readiness score each morning (hello, Whoop and Oura's entire business model) based on HRV, resting heart rate, sleep quality, and recent activity. Does this reinvent anything? No. Is it overdue? Absolutely.

Ceramic's Back, and I'm Conflicted

Remember the Series 2 and 3, when Apple offered a white ceramic case for people with more money than sense? It's back. Probably starts north of £1,200 based on past pricing, but we don't have official UK numbers yet. The ceramic looks gorgeous, scratches less than aluminium, and weighs about the same as stainless steel without the fingerprint magnet property.

Here's my issue: ceramic is brilliant until you drop it. Then it shatters. Aluminium dents. Steel scratches. Titanium laughs it off. Ceramic goes pop. I've seen it happen to a mate's Series 3 Edition when he clipped a doorframe wrong. Expensive paperweight.

That said, Apple's added Ceramic Shield 2 to all models, not just the ceramic case. The display is supposedly 50% tougher than last year. I'll believe it when I see long-term drop tests, but if it's anything like the iPhone's Ceramic Shield, it's a genuine improvement. My iPhone 14 has taken some nasty falls without a crack. Credit where it's due.

What's Missing (And Why It Matters)

Right. Deep breath. No blood glucose monitoring. None. Zero. We've had leaks about this for two years. The Apple Watch will "revolutionise diabetes care", they said. It'll happen when it happens, and Series 12 is not that day.

More frustrating: battery life is unchanged. Apple quotes "18-hour all-day battery" as if that's something to celebrate in 2026. My Garmin lasts two weeks. My Withings lasts a month. Even Samsung's managing two days now. Apple's stuck at "charge it every night or else", and for a watch pushing hardcore health tracking (including sleep analysis, which requires wearing it at night), that's absurd.

Fast charging helps. You can supposedly get 8 hours of use from 15 minutes on the charger. Fine. But you know what's better than fast charging? Not needing to charge as often. Apple has the engineering talent and the chip efficiency to pull off 36 hours minimum. They're choosing not to, probably because the Watch team reports to the iPhone team, and multi-day battery life means fewer chances to interact with your phone.

And Touch ID? Still missing. Face ID on your wrist makes zero sense, granted, but Touch ID in the crown would be brilliant for Apple Pay without needing to type a passcode. Apple owns the patent. They did it on iPads and MacBooks. Just do it here.

The 5G Question Nobody Asked

Series 12 supports 5G. So did Series 11. And Series 10. I've owned cellular Apple Watches since Series 4. I've used the cellular connection maybe a dozen times total. It's a nice safety net (GPS-only models lose functionality if you're away from your phone), but 5G specifically? On a watch? Why?

5G drains battery faster than LTE. It requires more complex antennas. It costs more to manufacture. And for what? Streaming music slightly faster? Downloading podcast episodes I could've synced at home? This feels like box-ticking for carrier partnerships rather than genuine user benefit.

Bluetooth 5.3 is a better upgrade. Lower latency for audio, better connection stability, improved range. I notice that daily with AirPods. I'd never notice 5G on my wrist unless I'm checking.

Should You Wait for This?

Goes on sale September 18. Starts at $399, which I'd guess translates to £399 in the UK because Apple's pricing team hates exchange rates. If you're on a Series 8 or newer, nothing here screams "upgrade now". On-device AI is neat, not essential. Ceramic is lovely if you trust yourself not to smash it.

If you're on a Series 6 or older, though, this is a solid jump. The health sensors are meaningfully better. The display is brighter. WatchOS keeps getting smarter (though it'll run on older models too, just slower). And frankly, if your current Watch's battery has degraded to the point where it dies by dinner, Series 12's "all-day" claim will feel like an improvement.

My take? Apple's playing it safe. They've added the features that Android watches already had (on-device processing, frequent HRV tracking, readiness scores), wrapped them in gorgeous hardware, and called it revolutionary. It's not. It's iterative. But iteration done well is still worth buying, especially if you're locked into the Apple ecosystem and use AirPods, iPhone, and Mac daily. The integration remains unmatched.

Just don't expect miracles. Expect a very good smartwatch that still can't last two days on a charge, still can't check your blood sugar, and still costs as much as a decent mechanical watch that'll outlive you. That's the Apple Watch in 2026. Better than ever, but not as much better as it should be.

The Bigger Picture

What frustrates me most about the Series 12 launch is how little it moves the conversation forward. Smartwatches in 2026 should be solving battery anxiety, health monitoring gaps, and over-reliance on smartphones. Instead, we're getting incremental processor bumps and material choices.

Garmin's pushing solar charging. Withings has a hybrid that lasts 30 days. Samsung's experimenting with health sensors Apple won't touch for regulatory reasons. And Apple? Apple's refining the same formula since Series 4, making it slightly better each year without taking real risks.

That's not necessarily wrong. The Apple Watch is the best-selling smartwatch by miles. Why mess with success? But for those of us who've been wearing these things for a decade, it's getting a bit samey. I want Apple to surprise me again, the way the original Watch did in 2015, or the Series 4 redesign did in 2018. Series 12 doesn't do that.

It's a good watch. It'll sell millions. But it's not the future of wearable tech. It's the present, slightly polished.