MacStories iOS 27 Review

One of my favorite things to do every September is sit down somewhere comfortable with a cup of coffee and read that year’s iOS review by Federico Viticci. While most of the Apple conversation this year has been about everything else (iPhone Duo, new CEO, etc), Federico spends his time on the software itself. How it looks, how it works, and what it actually means for the people using it.

The whole thing is worth your morning, but his take on the new Siri AI is the part I keep thinking about. He’s spent years building agentic workflows and really cool shortcuts at MacStories, so he’s better positioned than almost anyone to say where Siri falls short. Which obviously he does. And then he says something more interesting:

I don’t think Apple’s Foundation models are as good as GPT-5.5 Instant, but Siri AI is the more ergonomic and integrated product… If you’re a casual user of ChatGPT and do not pay for it, I think Siri AI can be a very compelling product.

I think the comparison most people are making is wrong. The question isn’t whether Siri beats the frontier models (it currently doesn’t and it probably won’t ever). The question is whether it’s good enough at the things people actually need it for. A model you can invoke from the Camera app or a context menu without switching apps is doing something a super-high tech and fast model stuck in a silo isn’t.

Can Siri replace Claude for the app and website work I do every day? Absolutely not. But it doesn’t have to, and that’s not what it’s for. Most people aren’t building software. Not everyone needs an F-150, some people just need a really nice Vespa.