Apple’s AI isn’t a letdown. AI is the letdown

Apple’s been taking heat lately for its AI efforts. And I get it. The delayed Siri upgrade, the comically useless text message summaries—it’s not a great look for a company that built its reputation on polish and reliability. But the narrative that Apple is somehow failing at AI misses the real story.

The truth is, AI is failing itself. Or rather, AI hasn’t actually delivered a product that justifies the hype. And Apple, for all its stumbles, might be the only company honest enough to admit it.

Let’s start with the obvious: why is everyone scrambling to shove AI into everything? Wall Street. Investors want a “super cycle” that drives upgrades. Consumers? They couldn’t care less. Apple’s own ad for AI got such a hostile reception they had to pull it. People don’t want this stuff. They want better battery life, a camera that works, and maybe a phone that doesn’t crack when you look at it wrong.

So Apple delayed its AI features. That’s not failure—that’s pragmatism. The company is saying, “This isn’t ready yet.” And honestly, that’s refreshing.

AI can only be failed

There’s a popular saying in policy circles: “The party can never fail, it can only be failed.” The same logic is creeping into AI discourse. AI can never fail—it can only be failed by you and me, the Luddites who don’t appreciate its potential.

Tech columnists like Kevin Roose at the New York Times have argued that Apple is failing AI. That Apple needs to be “more comfortable with error” and push out imperfect products. To which I say: absolutely not.

Apple’s entire brand is built on trust. It’s the reason people feel safe scanning their faces, storing bank info, and sharing their location. It’s the reason your parents can use an iPhone without reading a manual. That trust is hard-won and easily lost. Asking Apple to ship buggy, unreliable AI features isn’t visionary—it’s irresponsible.

And for what? What’s the killer use case? As Casey Newton pointed out on the same podcast, Google and Amazon haven’t figured it out either. Nobody’s rushing to buy a Pixel or an Echo because of AI. The tech is still a science project, not a product.

Large language models are fascinating. They’re an academic marvel with potential. But a bot that’s 80% accurate isn’t a consumer product—it’s a liability. And Apple, of all companies, knows that better than anyone.

So no, Apple isn’t failing AI. AI is failing to deliver something worth building on. And until that changes, I’d rather Apple take its time than rush out junk.