2025 will be remembered for what Apple didn’t deliver

Think back to the end of 2024. It was a more innocent time. Sure, after unveiling Apple Intelligence with great fanfare at WWDC in June that year, it hadn’t actually shipped much (other than a raft of TV commercials featuring nonexistent features). But surely the company wouldn’t let us down. 2025 was truly going to be the year Siri got fixed, and Apple Intelligence took flight.

Well, about that…

March 2025 arrived, and Apple admitted that it just couldn’t make it all work, and the more personalized Siri promised in 2024 wouldn’t come until sometime “in the coming year,” which, for the record, is next year.

Anyway, at the end of 2024, I made my usual predictions and wishcasting in public for Macworld–I’ve been doing this for, what, a decade now? And in those heady days, I trusted that Apple would deliver what it promised and that 2025 would be the “year of Apple Intelligence.”

I actually said that. I don’t know if I’ve missed a prediction as badly in the entire time I’ve been doing this. Apple had such a track record of not promising what it couldn’t deliver that I really did believe it would not promise a feature it couldn’t ship. But you know the story, as Apple tells it–the feature couldn’t meet the company’s “quality bar” and was therefore kicked into the following year.

Whatever else Apple did in 2025, and it actually did quite a lot, it will forever be remembered as the year that it admitted it had not lived up to its own quality bar when it came to shipping the features that it promised. And if failing to meet the AI moment ends up being an important data point in the long-term history of Apple, it will be easy to point directly at Apple’s 2025 rollback of 2024’s announcements.

Apple is a company that generally has avoided the trap of overpromising and underdelivering. That it happened shows just how much pressure Apple felt it was under to show that it had AI well in hand. The success reaped in 2024 came back to haunt it in 2025. I thought the company would just power through and get where it needed to go… and it simply couldn’t get there. It’s one of the biggest admissions of mortality and limitation that Apple has offered in more than a decade.

Still, by the end of last year, it was already clear that the Apple Intelligence rollout was more style than substance. In my predictions for the year, I said that Apple Intelligence would “continue to be the same mishmash of useful and useless features” that we had already seen, and that “by the end of 2025 Siri still won’t be as good as it should be.” The skepticism was right, but I certainly didn’t expect this level of failure.

Apple

Similarly, after Apple did all but declare Google Gemini as a future partner at WWDC 2024, we all expected Apple to announce that it was joining ChatGPT as an AI model connected with Siri on Apple devices. Apple executives would mention Gemini and essentially wink at the audience, again and again. And yet… it never happened. It still hasn’t happened. ChatGPT integration is all there is, and even that integration is so far from the current state of the art for ChatGPT. What a mess.

Apple’s failure to build a better Siri driven by AI had an effect on everything. I thought predicting Apple’s triumphant re-entry into the smart home market with a new product was almost a no-brainer. That turned out to be exactly the problem: Apple apparently designed a home-control device with smart Siri as its brain, and if there was no smart Siri, that product had to be put on hold. Nothing exposes the imbalance between Apple’s hardware designers and its software organization than multiple products reportedly being finished months or years in advance, forced to idle because their software isn’t up to snuff.

Fortunately, most of the time you bank on Apple’s hardware delivering, you get it right. After numerous reports that the company would ship the thinnest iPhone ever, it did that. I predicted that it would be priced less than $1,000, and since it starts at $999 I think I can say that I nailed it. I also predicted the iPhone Air would offer “enough technical compromises to infuriate a whole bunch of tech nerds.” That’s a prediction that combines my decades of knowledge of Apple with my decades of knowledge of tech nerds, and I think I pretty much nailed it.

Let’s face it: predicting the arrival of a bunch of M4 Macs and, eventually, the M5 was not so hard. I learned how to count when I was very young and I mostly don’t even need to use my fingers anymore. But nobody outside of Apple’s chip lab would’ve predicted that there would be a new M3 Mac in 2025, and there was, courtesy of the M3 Ultra-powered Mac Studio.

Foundry

Turning to Apple’s thriving services business, I expected continued growth–about as easy a bet as guessing that M5 would follow M4–as well as a big theatrical film and a bunch of prestigious TV shows. Between the arrival of F1 The Movie and the rise of awards darlings Severance and The Studio, Apple’s film and TV division had a great year.

But I also predicted that, despite all the buzz on the subject, Apple would end up not buying major sports rights. Instead, the company went out and outbid ESPN for Formula 1 rights in the United States. Guess all that time spent working with the racing circuit for the Brad Pitt film rubbed off.

Still, I correctly predicted that Apple wouldn’t try to buy a movie studio. While Netflix and Paramount continue to joust over control of Warner Bros. Discovery, which follows the Skydance purchase of Paramount, Apple seems content to not be involved with any of that. It’s got a pretty good thing going, I think.

Where does that leave us? Apple’s biggest surprise in 2025 was actually its failure to deliver what it promised in 2024, forcing the company to kick a big portion of Apple Intelligence down the road into 2026. Apple continued to release impressive new hardware products like clockwork in 2025, but its ill-advised promises at WWDC 2024 lingered like a hangover, darkening the vibe of the entire year.