News

Apple Tries Again to Make Siri Matter

• By John K. Waters
• 06/09/2026

For more than a decade, Siri has lived in the strange space between cultural ubiquity and technical disappointment. Everyone knew the voice. Fewer people trusted it to do much more than set a timer, start a call, or mishear a request.

Apple is now trying to change that.

At its Worldwide Developers Conference, the company introduced Siri AI, a redesigned version of its voice assistant that Apple describes in its own announcement as “a profoundly more capable and personal assistant.” The update is intended to make Siri more conversational, more context-aware, and more useful across iPhone, iPad, Mac, Apple Watch, and Vision Pro.

The announcement is less a product flourish than a course correction. Apple first promised a more personal, AI-powered Siri in 2024, but the company struggled to deliver the assistant it had described. The delay became one of the most visible signs that Apple, despite controlling the hardware and software used by hundreds of millions of people, had fallen behind in the generative AI race.

The new Siri AI is Apple’s latest answer to that problem. It's also an attempt to reframe the AI competition around an advantage Apple has long emphasized: the assistant already sits inside the operating system, close to the user’s apps, messages, calendar, photos, files, and device settings.

Apple says the new Siri is designed to understand more of a user’s personal context and what is happening on screen. It can respond in a more natural conversational style, act across apps, and handle requests that require a better understanding of the user’s device and data. Apple also introduced a dedicated Siri app with a chatbot-style interface, giving the assistant a more visible home beyond the familiar voice prompt.

The company is positioning the redesign as part of a broader expansion of Apple Intelligence, the AI system Apple has been embedding across its software platforms. Craig Federighi, the company’s senior vice president of Software Engineering, said in a statement the models that power Apple Intelligence are becoming “more capable and efficient,” and that Apple is integrating AI features “in even more places across each of our operating systems.”

That integration matters because Apple is not trying to build a chatbot that sits apart from the rest of computing. Its pitch is that AI should show up inside daily workflows: summarizing information, helping write, editing images, acting through apps, and understanding what users are trying to do without requiring them to jump into a separate AI product.

That strategy also carries risk. A system that promises to understand personal context needs access to sensitive information, and Apple has made privacy central to its AI messaging. The company says Siri AI queries will be processed either on device or through Apple’s Private Cloud Compute system, depending on the task. Federighi said the new Siri was designed “with privacy at every step.”

Apple’s approach contrasts with that of companies that have moved faster and more visibly into generative AI, including OpenAI, Google, and Anthropic. Those companies have defined much of the public conversation around chatbots, multimodal AI, coding assistants, and AI agents. Apple, by comparison, has tried to move more cautiously, saying its priority is to make AI useful inside its products rather than simply release new tools for their own sake.

At the same time, Apple’s caution has not insulated it from criticism. The company faced user frustration over the delayed Siri capabilities it had previously previewed. It also agreed in May to pay $250 million to settle a class-action lawsuit alleging it misrepresented Siri’s AI capabilities, without admitting wrongdoing.

The new Siri AI, then, is not just another software update. It's a test of whether Apple can translate its traditional strengths, hardware control, software integration, and privacy messaging into the AI era.

One of the most notable parts of the story is that Apple is not doing all of this alone. Multiple reports say the new Siri relies on technology developed in collaboration with Google, including Gemini-based models. Apple has framed that arrangement as a way to improve the underlying Apple Foundation Models while still preserving its own privacy architecture and product experience.

That partnership underscores how difficult the current AI race has become, even for Apple. The company has immense distribution, tight control over its ecosystem, and one of the world’s most valuable consumer brands. But frontier AI requires enormous model-development capacity, specialized infrastructure, and rapid iteration. Apple’s decision to work with Google suggests that, at least for now, the company sees outside model expertise as necessary to deliver the assistant it promised.

The rollout will not be universal at launch. Reports indicate that Siri AI will begin with limited language and regional availability, with English support first. Apple has also said regulatory issues will affect initial availability in some markets, including the European Union.

For users, the practical question is simple: Will Siri finally become useful enough to change behavior?

That is the bar Apple now has to clear. The company does not need Siri AI to win every benchmark or look like every other chatbot. It needs the assistant to work reliably in the small, repetitive, context-heavy moments where people actually use their devices.

If it does, Apple could turn Siri from a long-running punchline into the default AI interface for millions of people who may never download a separate chatbot app. If it does not, the new branding will only make the gap between promise and performance harder to ignore.