WWDC 2026: Apple Intelligence & Siri AI with Gemini

Apple has dropped a very big WWDC 2026 announcement that blindsided nearly everyone who has been following its keynotes for years.

At the heart of the show-and-tell is the rebranded Siri, now called Siri AI. Apple Intelligence also saw massive changes. Why? That’s because it now has Google Gemini backing both of them.

Here’s what really went down at WWDC 2026. Fair warning, it’s a lot, and we’ve only plucked out the game-changing details.

New architecture for Apple Intelligence

Apple is now working with Google, using Gemini to create integrated Apple Intelligence on-device models for more advanced reasoning across additional modalities (beyond text and photos).

Also improved is Apple Intelligence’s ability to work with the user’s speech, with better dictation and improved understanding across apps, texts, and images.

The new Apple Intelligence model also adds a System Orchestrator, adding more personal context understanding across the system’s apps and background information. This unlocks Apple Intelligence system-wide and also enables it to pull, use, and parse personal information to help the user clear away multi-step tasks on their iDevice.

Also added is Broad World Knowledge, so Apple Intelligence is also not confined to the information on your device. It can now take in real-world updates to provide answers, sourced from the Internet.

Apple said that the new models are both on-device and rely on Private Cloud Compute to process AI requests. Data and information processed on Private Cloud Compute is only accessible to the user, and Apple claims that it doesn’t store or access that information.

Siri + Apple Intelligence + Gemini? = Siri AI

Siri is rebranded as Siri AI because it now incorporates Apple Intelligence. But that’s not the only trick up its sleeve, as Apple has done the unthinkable: working with Google to incorporate Gemini features into Siri AI.

App actions, on-screen interaction, and broad world knowledge (the hallmarks of the improved Apple Intelligence with Gemini) are all included with the new Siri.

Some examples given (that’s also new) is Siri AI’s ability to look up new concerts that are happening, identify the concert tickets’ sales method, and set a reminder to buy.

It can also look up the contents of the screen and answer questions based on what’s shown. Additionally, it can dig into the phone’s contacts to suss out details like address and route, or dig into Photos and share relevant ones in Shared Albums.

Additional features also extend to how Siri AI presents itself, with voice adjustments now including pace and expressivity.

Siri AI gets “in-depth answers”, which is very similar to how Gemini operates, with more contextual awareness and more natural responses. This lets Siri AI create in-depth plans, give deeper feedback, and be more verbose beyond one-question-one-answer requests. For iPhones, you can say “Hey Siri” or swipe down from the Dynamic Island to ask away.

For the Mac, Siri AI is integrated into its search function, Spotlight. Ctrl+Click triggers this command. It can also analyse files on the Mac and help with trawling through files and documents.

Aiding the entire spread of Siri AI across Apple’s ecosystem is a dedicated app. It allows users to go back and dig out past Siri AI conversations on iPadOS, MacOS, and iOS, and even take questions on watchOS. For visionOS, the Siri App works even when the user simply looks at the app icon.

Siri AI’s visual understanding has also seen a massive upgrade. iPhones also get a new Siri Mode in the Camera app. It suggests relevant actions based on what the camera is pointed at and where, so there’s no need to take photos or screenshots to ask about what’s shown.

For macOS, using a dedicated keyboard shortcut lets users select a section on their display, before asking Siri AI for an answer. For visionOS, it goes even further by supporting questions simply by looking at objects or app content.

Writing Tools, the assistive writing feature, also got upgraded with Siri AI, and it cuts across your writing across different apps (Messages, Mail, etc.) and gives automatic proofreading.

Siri AI now supports the English language, and Apple said it will “quickly” expand to include more languages.

Using Apple Intelligence in other apps

Safari, the default Apple web browsing app, gets improved tab management. New to the app is Organise, which uses Apple Intelligence to bring related tabs together, and users can close the whole tab group or save it for later.

Safari can also automatically monitor a webpage with Notify Me, which sends a notification if the webpage’s contents change (really handy for waiting on products that are out of stock).

Additionally, users can create browser extensions by simply describing what they need to Safari.

Passwords allows users to fix passwords across all their websites, but Apple Intelligence lets users do so directly via the app instead of going into individual websites to update them (only for eligible accounts).

Messages get contextual understanding and one-tap suggestions, like reminders or note creation. These suggestions also cut across Mail, Phone, to surface helpful information during critical moments like calls.

For example, Calendar also creates and saves events, but Apple Intelligence lets you just recite the appointment so the fields save the information (time, date, type of appointment, etc.).

In the Home app, a bundle of notifications can now be recognised as one continuous activity, so it does not bombard the user each time a door is open or a smart light is switched on. It also reviews multiple security camera feeds and watches recorded clips to summarise and extract key moments, such as what the dog did and the exact frames when the package was delivered.

Shortcuts, which simplify multi-step tasks across apps, now support natural-language descriptions where you can create one by describing. This spares the user from manually creating a shortcut by pulling together bits and bots across the system and app actions you need to automate a behaviour.

Image Playground, the generative AI visual app, adds photorealistic support and natural-language add-ons. Adding, subtracting, or changing an entire image’s vibe can be done by describing the change. It can also create new scenes based on the people found inside your Photos app.

The Photos app’s image editing tools also get the Apple Intelligence treatment. The new addition is Spatial Reframing, which allows users to manipulate a photo’s composition. Tapping “Reframe” when editing a photo lets users tap or drag to shift perspective and preview edit effects before committing the change.

Goodness, when will these features arrive and are they free-of-charge?

While Siri AI is only supporting English for now, Apple Intelligence is supported on existing languages that already have the old version:English, Danish, Dutch, French, German, Italian, Norwegian, Portuguese, Spanish, Swedish, Turkish, Vietnamese, Chinese (Simplified), Chinese (traditional), Japanese, and Korean.

Apple Intelligence and Siri Ai goes to iOS 27, iPadOS 27, watchOS 27, and visionOS 27.

However, Apple Intelligence only starts on iPhone 16 models or later, iPhone 15 Pro and Pro max, iPad mini (A17 Pro), iPads with M1 chip or newer, Apple Vision Pro, Apple Watch Series 10 or later, the Apple Watch Ultra 2 or later, and the Apple Watch SE 3 if paired with an Apple Intelligence-enabled iPhone nearby.

Siri AI goes to iPhone Air, iPhone 17 Pro series, iPad (M4 chip), and Mac (M3 chip) first.

Features like image generation have daily usage limits, which can be increased with an iCloud+ subscription plan.

The device OS start their developer beta today, and officially go live for users in Fall 2026.

For more on WWDC 2026, check out the articles below.