Apple Considers Google for AI Siri

This potential partnership follows reports that Apple's own large language models were not performing well enough in testing, prompting the company to explore collaborations with other AI leaders, including OpenAI and Anthropic. The deal with Google would be structured as a cloud-computing contract rather than a direct licensing of the technology. Under the reported multi-year deal, Apple would gain access to Google's Gemini AI models to serve as the foundation for new features in Siri and across "Apple Intelligence," the company's suite of AI tools. Apple would then build its own systems on top of Google's foundational layer, controlling the user interface and ensuring integration with its ecosystem and privacy standards. This move highlights a significant strategy divergence in the AI "arms race." While competitors like Google, Microsoft, and Meta are investing tens of billions annually in AI infrastructure, Apple's planned spending is considerably lower, estimated at just over $14 billion for fiscal 2026. Apple is opting for a hybrid approach, combining its own on-device processing with cloud services from partners like Google. Google, in contrast, is massively scaling its AI infrastructure, with plans to double its AI serving capacity every six months. The company is developing its seventh-generation custom AI accelerators, known as Tensor Processing Units (TPUs), to power its services. This infrastructure not only supports Google's own products but is also a key part of its cloud offerings. For Apple, this collaboration offers a way to accelerate the rollout of advanced AI features for Siri, which has been long-promised but delayed. Analysts suggest that by leveraging Google's mature technology, Apple can reduce its time-to-market and execution risk. The financial arrangement could see Apple paying Google around $1 billion annually for the use of its Gemini models. This is a notable reversal of their long-standing search deal, where Google pays Apple billions to be the default search engine on iPhones. Apple's broader AI strategy includes a significant four-year, $500-billion investment in U.S.-based projects, including a new server manufacturing facility in Texas and expanded data center capacity. This indicates a long-term goal of vertical integration, even as it currently relies on partners for heavy-duty AI processing.