Core Insights - Apple is in discussions with Google to use Google's Gemini AI as the core technology for the next generation of Siri [1][4] - The talks are in the early stages, but Apple has shown a proactive approach by reaching out to Google for a customized AI model for Siri [4] - Google has begun training a model that can run on Apple's private cloud servers, indicating the importance of this collaboration [4] Collaboration Strategy - Apple is not only engaging with Google but has also previously discussed with OpenAI and Anthropic for developing models for Siri [4] - This approach reflects Apple's strategy of exploring multiple partnerships to find the most suitable AI technology for Siri [4] Internal Development - Despite seeking external collaborations, Apple is also testing several large language models (LLMs), including its own, to determine which provides the best consumer experience [4] - Two versions of the new Siri are under development: one using Apple's own model and another utilizing a third-party model [4] Timeline - The upgraded version of Siri, which will incorporate large language models, is expected to be launched in the spring of 2026 [4]
苹果为Siri升级广撒网,谷歌Gemini AI或成关键“拼图”