Apple's AI Siri relaunch is the most predictable thing at WWDC 2026

I've seen this movie before, by the way. This is the self-driving car hype cycle all over again, where every year is supposedly the year everything changes, and then you look up five years later and we're still talking about "next year." Tesla promised full autonomy in 2017. Waymo's been "almost there" since 2018. And now Apple's been promising smart Siri since 2024.

Call me old-fashioned, but I think you should probably ship the thing before you announce it three times.

Playing from behind (on purpose?)

Here's where it gets interesting, and where I'll admit some genuine uncertainty about what Apple's doing.

The Verge makes a point I hadn't fully considered: playing from behind might not be such a bad move. While Google and OpenAI and Anthropic have been sprinting ahead, burning cash, and occasionally embarrassing themselves with AI that hallucinates or goes off the rails, Apple's been... waiting. Watching. Letting other companies make the expensive mistakes.

There's a version of this story where Apple's caution looks smart in retrospect. Where they let the market mature, let the models get better and cheaper, let users figure out what they actually want from AI assistants (turns out it's not philosophical debates with Claude, it's "remind me to buy milk when I'm near Trader Joe's"). Then Apple swoops in with something polished and practical while everyone else is still explaining why their chatbot told someone to eat rocks.

I don't know if that's what's happening here. It's too early to say whether this third attempt at AI Siri will actually deliver. But I'll give Apple this: the timing is interesting.

The AI market is, to use Bloomberg's framing, "down." Investor enthusiasm has cooled. The companies that went all-in on AI are facing hard questions about monetization and whether consumers actually want this stuff. Apple, which generates roughly $380 billion in annual revenue from hardware and services, can afford to be patient in ways that AI-first startups cannot.

So what

The question I keep coming back to is whether any of this matters to actual users. Not developers, not investors, not tech journalists like me who get paid to have opinions about product announcements. Regular people.

My wife uses Siri exactly twice a day: to set timers while cooking and to add things to her shopping list. She doesn't care about chatbot interfaces or deeper integration or whatever else Apple's promising. She cares about whether the timer actually gets set, which it does, and whether the shopping list syncs to her phone, which it mostly does.

Maybe that's the whole game here. Maybe Apple knows that most people don't need Siri to be ChatGPT, they just need it to be reliable and private and integrated into the stuff they already use. And maybe all this AI theater is really about convincing developers to build for the platform and investors to keep buying the stock.

But what do I know. I still prefer email to Slack and I think the best software interface ever designed was the 2007 iPod click wheel. I'm probably not Apple's target demo for AI features.

What I do know is that we've now had three WWDC keynotes promising us the AI-powered Siri of the future. At some point, the future has to actually arrive. This year, Apple's promising that the chatbot-style interface and cross-device integration are real, they're shipping, they're not "coming later this year" with an asterisk.

I'll believe it when I see it working on my phone. And even then, I'll probably still just use it to set timers.

If you want to argue about this, my email's on the about page. I read every one, eventually.

Sarah Williams · Yesterday · 4 min