He put AU$5,999 of Apple's future onto his 86-year-old grandmother, and she loved it. She figured out the eye-tracking in seconds, flipped through photos, then got drawn into an immersive video of a person riding a bull and was, in the reviewer's words, "somewhat terrified by the flight scene that followed."
It's a lovely moment. It's also the moment a reviewer called the M5 Vision Pro "a wildly expensive, slightly flawed look into the future of technology."
He wrote that in a review published three months into ownership. Roughly the same window, Apple's incoming CEO quietly deleted that future from the roadmap.
Last week in this space, Ternus scrapping the Vision Pro was still a rumor worth flagging. This week it has a shape: two glasses, one in 2027, one in 2029, and nothing with a head strap in between.
Here is the gut-punch for anyone who bought in. You now own the bridge. And the man about to run Apple just confirmed there is nothing on the other side.
You now own the bridge. And the man about to run Apple just confirmed there is nothing on the other side.
Start with what changed on paper. In June 2025, analyst Ming-Chi Kuo published a Vision roadmap with seven head-mounted products. In June 2026 he published an update with two. Both are glasses. Neither is a headset.
Kuo says the overhaul was "signed off by Apple's next CEO, John Ternus," who takes over September 1. This is the first major product call we can pin directly on the incoming chief executive, and he used it to cut everything Apple spent a decade and a reported multibillion-dollar budget building toward.
The second-generation Vision Pro: gone. The lighter, cheaper Vision Air that was supposed to widen the market: gone, killed off back in October 2025 per Gurman. The Mac-tethered display glasses: sunset in January 2025.
Kuo's verdict on the cull is blunt: "I think removing the Vision Pro line was the right call, as Apple shifts resources toward smart glasses with greater mass-market potential."
Two independent tracks now converge on the same finding. Gurman from Bloomberg, Kuo from the Asian supply chain, both agree there is no next Vision headset shipping. The only daylight is whether you call it dormant or dead — Gurman says a Vision Pro 2 is "in testing" but the category is "on ice." That's a semantic argument, not a strategy.
Now here's the strange thing about that news, and it's the part most coverage misses. For a buyer, cancellation is clarifying.
The usual Apple purchase is a war with your own patience: buy now, or wait eight months for the better one? That war is over. The M5 is the ceiling. There is no next one to wait for, no successor to feel dumb for missing, no "should have held out."
If you want a Vision Pro, the question collapses to a single line: do you want the thing it is right now? Because the thing it is right now is all it will ever be.
So what is it right now? The review is honest to the point of self-incrimination. The M5 chip fixed the processing stutters; the Dual Knit Band fixed the neck cramps. It packs a 32:9 ultrawide curved monitor and a 100-inch personal 3D cinema into a case that fits in your carry-on.
And then the reviewer admits he mostly uses it as an 80-inch screen for the Sunday footy in the middle of his living room. His own words: "you can surely purchase a pretty impressive TV for less."
That's the whole verdict in one sentence. This is a luxury travel monitor and a private cinema. It is genuinely, absurdly good at being those two things. The battery still dies in about 2.5 to 3 hours, and its use cases in the home, the reviewer concedes, "remain a mystery."
Buy it for that. Buy it if you fly long-haul constantly and want the best portable screen money can build. Buy it if you're a creative pro who wants an ultrawide desk that folds into luggage. Do not buy it because you think it grows into something. Ternus already told you, in one signature, that it becomes nothing.
Which brings us to the harder question — whether the glasses bet Apple traded it for is actually a vision, or just a spreadsheet.
Look at the numbers driving the pivot. Global smart-glasses shipments grew 139% year-over-year in the second half of 2025, per Counterpoint Research. Meta held 82% of that market in the same period.
Against that, Apple watched a $3,499 headset undershoot expectations twice — once at launch, again with the M5 refresh — while Meta sold Ray-Bans by the millions at a tenth of the price.
That's not a founder staring into the future. That's an executive reading a sales chart. Kuo says as much when he frames the cut as chasing "mass-market potential." Ternus isn't betting glasses are the answer; he's betting headsets are the wrong question.
But conviction and desperation can look identical on a roadmap, and there are two tells that this is closer to the latter.
The first is timing. Apple's AI glasses aren't expected until the end of 2027. That reportedly hands Meta another eighteen months to sell, iterate, and build the retail infrastructure that makes smart glasses feel normal. Apple's playbook — enter late, win on brand and iPhone integration, the way it did to smartwatches in 2015 — only works if its first pair is as good as or better than whatever Meta ships by then. That's a bet on execution, against a company with an 82% head start.
The second tell is more damning: Apple isn't even sure glasses are the answer it wants. It's reportedly working on AirPods Pro with built-in cameras — a hedge that delivers the AI-assistant experience without the glasses part. Plus the recurring pendant and pin rumors.
Read that honestly and the pattern is clear. Apple is chasing "AI you can wear," and glasses are merely the current favorite guess. The company that just cancelled its headset because it never solved the buying question is now placing multiple, mutually competing bets on the next form factor.
That's the real reason the Vision Pro died. Not because glasses are inevitable — but because after two undershoots, Apple still couldn't answer the only question that matters: who is this for, and why now? A displayless camera on your face at a Ray-Ban price at least has an answer. A $3,499 headset never did.
So here is the verdict, and it hasn't softened.
If you already own the M5 Vision Pro, you own a fully supported, fully orphaned device. Software looks solid through 2028 or 2029 — visionOS 27 is expected to bring the new Siri and Photos AI tools — so run it on a depreciation schedule that assumes the hardware you bought is the hardware you keep. That's fine for a private cinema. It's a harder pill for the hospitals and training departments that bought in on an assumed long roadmap.
If you're a would-be buyer, the calculus is now brutally simple. There is nothing to wait for, so buy it only if the 100-inch cinema and the carry-on ultrawide are worth AU$5,999 to you today.
Never buy it for what it will become. As VR.org put it, bridges are useful. They're also things you cross once.