Apple WWDC 2026 Preview: Gemini-Powered Siri and iOS 27 Arrive June 8

Apple’s Worldwide Developers Conference 2026 opens on June 8 with a keynote at 10 a.m. Pacific Time, and for the first time in years the company’s most anticipated announcement isn’t a new chip or a new form factor — it’s software. Specifically, it’s a rebuilt Siri that leans heavily on Google’s Gemini AI models, a third-party AI extension system that could fundamentally change how Apple Intelligence works, and a complete refresh of every operating system in Apple’s lineup.

The stakes are unusually high. Apple’s AI ambitions have repeatedly underdelivered since the original Apple Intelligence announcement at WWDC 2024. Siri’s promised “personal context” features shipped late, the upgraded ChatGPT integration was underwhelming compared to competitors, and a class action lawsuit — recently settled, with some iPhone 16 buyers eligible for up to $95 — alleged that Apple had misled consumers about the AI capabilities of its devices. Two years after making AI a flagship narrative, Apple needs a credible answer to the question it has been unable to satisfactorily answer: what does Apple Intelligence actually do that competitors don’t?

WWDC 2026 is that answer — or it’s nothing.

The Gemini Partnership: A Surprising Bet

The most significant structural change coming to Apple Intelligence is a deep integration with Google’s Gemini AI models. Apple and Google have maintained a complex, occasionally combative relationship for years — Google pays Apple billions annually to remain the default search engine in Safari — but the Gemini partnership represents something qualitatively different: Apple delegating the core generative intelligence of its flagship AI product to a competitor’s model.

The arrangement, described by multiple sources as a multi-year agreement, sees Gemini models serving as the generative backbone for a rebuilt Siri experience on iOS 27. The new Siri will handle open-ended, conversational queries in a way the current version fundamentally cannot: real-time web search, reasoning over documents and images, generating written content, and maintaining multi-turn context across a conversation rather than resetting with every query.

The result, according to early descriptions, feels “closer to a chatbot than the voice assistant users have known for years” — a deliberate shift in positioning that acknowledges what users have actually been asking Siri to do, and haven’t been able to.

A Third-Party AI Extension System

Potentially as important as the Gemini integration is a new extension-style system that, for the first time, would allow users to choose among different AI models for different tasks within Apple Intelligence. Sources indicate the system would allow third-party AI assistants — including Claude, ChatGPT, and potentially others — to integrate more deeply with Apple Intelligence features like Writing Tools and Image Playground, rather than being siloed in their own separate app experiences.

If Apple ships this as described, it would be a meaningful reversal of the company’s historical tendency to control AI layers tightly. It would also create a new competitive dimension within the Apple ecosystem: instead of fighting for share of a user’s home screen, AI companies would compete to become the preferred AI assistant within Apple Intelligence itself.

The business logic is clear. Apple cannot develop and maintain a frontier AI model at competitive cadence — training runs of the scale required by GPT-5, Gemini 3.5, or Claude 4 require infrastructure investments and research organizations that Apple, despite its trillion-dollar cash reserve, has not built. Partnering with leaders in each capability vertical, while controlling the interface layer, hardware integration, and privacy architecture, is Apple’s realistic path to an AI-native experience.

What iOS 27 Is Expected to Bring

Beyond Siri, the iOS 27 release is expected to introduce a range of AI-augmented features across the standard app suite:

Real-time web search in Siri: Rather than answering from a static knowledge base, the new Siri will pull live information from the web and synthesize it into conversational responses — closing one of the most glaring gaps between Siri and competitors like ChatGPT and Perplexity.

AI-powered image analysis and generation: Enhanced Image Playground is expected to support more sophisticated image editing workflows, including document analysis via camera, visual search improvements, and new generation capabilities.

File uploads and document reasoning: A consistent complaint about Siri has been its inability to reason over files, PDFs, and documents the way ChatGPT or Claude can. iOS 27 is expected to address this with support for file uploads in Siri conversations.

Cross-app context: Apple’s “Personal Context” vision — Siri understanding and acting across apps using on-device data — has been partially available since iOS 18.1 but limited in scope. iOS 27 is expected to significantly expand the range of apps and actions Siri can intelligently navigate.

The full platform update — iOS 27, iPadOS 27, macOS 27 (which has generated speculation about a new naming convention), watchOS 27, tvOS 27, and visionOS 27 — will be previewed at the keynote. Developer betas will be available the same day; public betas are expected in July, with general release in September.

The Hardware Question

WWDC is traditionally a software-focused event, but speculation persists about hardware announcements. The most credible rumor involves an updated Mac lineup refreshed with Apple’s next-generation chip, though Apple has not confirmed any hardware plans.

More interesting is the ongoing evolution of Apple’s augmented reality ambitions. With Google having shipped its Android XR glasses via Samsung at the beginning of the year and generating significant developer momentum, Apple faces pressure to show some tangible progress on wearable AR — even if a full consumer product is still years away.

Why This WWDC Matters More Than Most

Apple holds roughly 27% of the global smartphone market, a figure that understates its influence among premium users and developers. Whatever Apple ships in iOS 27 will be on over a billion active devices within 18 months of release. That scale gives Apple’s AI decisions a weight that few announcements in technology match.

The Gemini partnership, if it performs as described, would give hundreds of millions of users their first hands-on experience with frontier-class generative AI in a consumer product they already trust and use daily. For Google, distributing Gemini through Apple’s platform is a major go-to-market win. For Apple, it’s a bet that controlling the experience layer — the UX, the privacy guarantees, the hardware integration — matters more than which company trained the underlying model.

Whether that bet is right may be the defining question of Apple’s next decade.

The keynote begins June 8 at 10 a.m. PT, streaming live on apple.com.