Did Apple Intelligence flop?

However much Apple CEO Tim Cook insists the company hasn’t fallen behind in the artificial intelligence race, it’s hard not to notice: Apple arrived last to the game.

A confusing debut

Announced at WWDC24, Apple Intelligence came first to the iPhone 16 — but only through the iOS 18.1 update. Many people bought the phone expecting AI out of the box and found out it would only be released months later.

That may have been the first mistake: creating an expectation the launch didn’t meet. Apple Intelligence has advanced features, but in everyday use few of them really make a difference. Here is what worked and what didn’t.

What worked

• Clean Up: removes unwanted objects from photos, as Google’s Pixel phones and Samsung’s Object Eraser already did. Before, in Apple’s ecosystem, you had to turn to third-party apps.
• Visual Intelligence: exclusive to the iPhone 16, it translates text and recognises plants and objects. It is triggered by the phone’s Camera Control button — which finally earns its place. It works like Google Lens; curious that Apple took so long to build something like this into iOS, and it may not even have needed AI.
• Spotlight and Siri on macOS: native integration makes AI accessible from many parts of the system. Spotlight became more complete and Siri more intuitive — with plenty of room to improve.

What didn’t

• Languages: at the time of writing, Apple Intelligence still doesn’t support Brazilian Portuguese. If the device isn’t set to English, the features simply don’t appear — a sign the product is still in training, while the competition already speaks many languages.
• Image Playground: the image creation app seems designed for children. The features are basic and the images rarely have the quality for real use. If you think Midjourney struggles with fingers, teeth and hair, Apple Intelligence does far worse.
• macOS integration: the system has Automator, which automates tasks such as building presentations, basic image editing, batch file renaming and conversions. It was disappointing to see that Apple Intelligence doesn’t talk to it yet — it would be a huge productivity gain for repetitive tasks.
• Lots of marketing, little delivery: the announcement made a big splash and set expectations of a category leader. That is not what happened. The aggressiveness of the marketing still doesn’t match the minimum quality expected for everyday use.

Verdict

Training and refining an AI takes time. But, so as not to fall behind the competition, Apple may have put an unfinished product on the market.

For now, Apple Intelligence is far from making a real difference in the lives of iPhone and Mac users. Still, integrating AI into the operating system has enormous potential, especially to make productivity simpler and more centralised. We are still waiting for a broader — and more useful — version 2.0.