Apple is reportedly working to distill Google's multi-trillion parameter Gemini AI to run on iPhones, according to The Information. The company has delayed its AI-enhanced Siri multiple times since first promising it in 2024, but a deal with Google will merge the iconic assistant with Gemini later this year. As the Worldwide Developers Conference approaches, Apple has been focusing on bringing advanced AI capabilities to smartphones, which have limited processing power. While Apple has long emphasized privacy by running AI locally, a new report suggests that the iPhone's Gemini-powered Siri will rely heavily on Google and Nvidia in the cloud. The Information reports that Apple's Gemini-infused Siri will run both on-device and in the cloud, a shift from its previous privacy-focused approach. Despite advancements in chip design, smartphones still struggle to handle large AI models due to limited RAM. The Information notes that the largest AI models are still basic assistants, making local AI challenging. On-device AI models are also quantized to run at lower precision, which affects token generation accuracy. Google's Gemini models, with trillions of parameters, are significantly larger than the smaller AI models that run on phones, which have at most a few billion parameters. The Information reports that Apple has struggled to run Google's undistilled Gemini models on its Private Cloud Compute infrastructure, which uses M-series Mac chips. Apple has reportedly signed a deal with Nvidia to use its Confidential Computing platform for cloud processing, which keeps data encrypted on Nvidia GPUs. This could help Apple maintain its privacy claims while routing complex tasks to Google's cloud infrastructure. *Source: [arstechnica](https://arstechnica.com/ai/2026/05/apple-reportedly-trying-to-distill-googles-multi-trillion-parameter-gemini-ai-to-run-on-iphone/)*