Apple plans to integrate a customized model of Google's Gemini AI into Siri, according to a Bloomberg report.

This collaboration could cost Apple around 1 billion dollars a year for access to technology that will manage complex tasks such as generating summaries and planning functions. Apple Private Cloud Compute servers will host this special version of Gemini, ensuring user data security. Additionally, Apple's internal AI models will continue to support other functions of Siri. The new Gemini model will operate on 1.2 trillion parameters, a significant increase from the 150 billion of current models.

This collaboration comes after months of internal discussions at Apple regarding the acceleration of AI feature launches for Siri. Although Apple evaluated partnerships with OpenAI and Anthropic, it chose Google as the main partner. However, the integration of the Gemini chatbot into Siri is not included in this agreement. Apple's CEO, Tim Cook, confirmed that the new Siri experience is expected to be launched next spring, emphasizing that Apple remains open to integrating other AI solutions from third parties. Despite its reliance on Google's technology, Apple continues to develop its own large-scale AI systems, aiming to replace Gemini in the long term.