2025 will be remembered for what Apple didn’t deliver

Think back to the end of 2024. It was a more innocent time. Of course, after unveiling Apple Intelligence to great fanfare at WWDC in June of that year, it hadn’t really delivered much (apart from a series of TV commercials showcasing non-existent features). But the company surely won’t let us down. 2025 was truly going to be the year Siri was fixed and Apple Intelligence took off.

Well, about that…

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

Anyway, at the end of 2024, I made my usual predictions and wishes in public for Macworld – I’ve been doing this for, what, a decade now? And in those heady times, I had confidence that Apple would deliver on its promises and that 2025 would be “the year of Apple intelligence.”

Apple

Actually, I said that. I don’t know if I’ve missed a prediction that much in the entire time I’ve been doing this. Apple had such a habit of not promising what it couldn’t deliver that I really thought it wouldn’t promise a feature it couldn’t deliver. But you know the story, as Apple tells it: the feature couldn’t meet the company’s “quality bar” and so was released the following year.

Whatever Apple did in 2025, and it did a lot, this year will always be remembered as the year it admitted it hadn’t met its own quality bar when it came to delivering promised features. And if AI’s failure to respond to timing ends up being an important data point in Apple’s long-term story, it will be easy to point directly to Apple’s 2025 rollback from 2024 announcements.

Apple is a company that has generally avoided the trap of over-promising and under-delivering. That this happened shows how much pressure Apple felt to show that it had AI in hand. The success it had in 2024 came back to haunt it in 2025. I thought the company was just going to get through it and get where it needed to go… and it just couldn’t get there. This is one of the biggest admissions of mortality and limitation offered by Apple in over a decade.

Yet 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 hodgepodge of useful and useless features” that we’ve already seen, and that “by the end of 2025, Siri still won’t be as good as it should be.” The skepticism was justified, but I certainly didn’t expect this level of failure.

Apple

Likewise, after Apple virtually declared Google Gemini as a future partner at WWDC 2024, we all expected Apple to announce that it was joining ChatGPT as a Siri-connected AI model on Apple devices. Apple executives would mention Gemini and essentially wink at the audience, over and over again. And yet… that never happened. He always that didn’t happen. The 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 create a better AI-driven Siri has 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 the brains, and if there was no smart Siri, that product had to be put on hold. Nothing reveals the imbalance between Apple’s hardware designers and its software organization than multiple products seemingly finished months or years in advance, forced to sit idle because their software isn’t up to par.

Fortunately, most of the time you bank on delivering Apple hardware, you succeed. After numerous reports that the company would ship the thinnest iPhone ever, it did so. I predicted it would be priced under $1,000, and since it starts at $999, I think I can say I succeeded. I also predicted that the iPhone Air would offer “enough technical compromises to infuriate an entire group of tech enthusiasts.” It’s a prediction that combines my decades of Apple knowledge with my decades of tech nerd knowledge, and I think I nailed it.

Let’s be real: predicting the arrival of a bunch of Mac M4s and, ultimately, the M5 wasn’t that difficult. I learned to count at a very young age and I hardly need to use my fingers anymore. But no one outside of Apple’s chip lab would have predicted there would be a new M3Mac in 2025, and there was, thanks to the Mac Studio equipped with M3 Ultra.

Foundry

When it came to Apple’s thriving services business, I expected continued growth — a bet as easy as guessing that M5 would follow M4 — as well as a big theatrical movie and a bunch of prestigious TV shows. Between the arrival of F1 The Movie and the rise of cherished rewards Breakup And The WorkshopApple’s film and television division had a great year.

But I also predicted that, despite all the buzz on the subject, Apple would eventually not buy major sports rights. Instead, the company outbid ESPN for the rights to Formula 1 in the United States. Guess all that time spent working on the race track for the Brad Pitt movie has rubbed off.

Yet I correctly predicted that Apple wouldn’t I try buying a movie studio. As Netflix and Paramount continue to vie for control of Warner Bros. Discovery, following the acquisition of Paramount by Skydance, Apple seems content with not be involved in all of this. It’s a very good thing, I think.

Where does this leave us? Apple’s biggest surprise in 2025 was actually its failure to deliver on its promises in 2024, forcing the company to give up much of Apple’s intelligence until 2026. Apple continued to release impressive new hardware products like Clockwork in 2025, but its misguided promises at WWDC 2024 remained like a hangover, darkening the mood of any the year.