According to Mark Gurman, Apple has finalized its strategy for a significant Siri update, set to debut with the release of iOS 26.4 in the spring of next year. This update promises to enhance the Siri experience by integrating advanced technology, specifically utilizing Google Gemini models. These custom models will be hosted on Apple’s Private Cloud Compute servers, enabling Siri to respond more effectively to user inquiries.
The revamped Siri will be capable of answering personalized questions, such as “find the book recommendation from Mom.” This functionality will be achieved by intelligently searching through data stored on your device and generating appropriate responses in real-time. As previously reported by Bloomberg, the new architecture of Siri will consist of three key components: a query planner, a knowledge search system, and a summarizer. The integration of Google’s Gemini models will enhance both the planner and summarizer functions, significantly improving Siri's performance.
Apple is committed to maintaining user privacy. The Google models will operate solely on Apple’s server infrastructure, ensuring that no external data is shared. Additionally, on-device personal data is expected to be processed using Apple’s own Foundation Models. This approach not only safeguards user information but also enhances the intelligence of Siri.
The new knowledge search component, potentially driven by Gemini models, will allow Siri to possess a deeper understanding of various world topics and trivia. This capability will enable the voice assistant to answer general knowledge questions more effectively, avoiding reliance on third-party integrations like the less efficient ChatGPT integration, or basic web search results. This upgrade aims to make Siri a more competent and reliable virtual assistant.
While the backend implementation of Gemini models is expected to be extensive, Bloomberg notes that Apple is unlikely to promote this partnership overtly. Instead, the features powered by Gemini will be seamlessly integrated into Siri, marketed as proprietary Apple technology running on Apple servers and presented through a familiar Apple user interface. This strategic decision allows Apple to address essential technology gaps where its own large language model (LLM) systems currently fall short.
This partnership resembles the approach taken by Samsung with its Galaxy phones, where many of the AI features are built upon Google Gemini capabilities. Ultimately, this arrangement provides a practical solution, enabling Apple to deliver the Siri AI features that it initially promised to its users over a year ago. Customers primarily seek enhanced functionality and features on their devices, rather than delving into the technical intricacies of implementation.
Apple is expected to announce the new Siri features in conjunction with the launch of iOS 26.4 in March or April. Additionally, the company is reportedly developing a new smart home display device that will leverage the advanced capabilities of the new AI assistant. This upcoming release is set to redefine user interaction with the Siri platform, making it a more valuable tool in everyday life.
