Apple Siri AI: Google Partnership & Frontier Models Explained

Apple has unveiled its revamped Siri, now called Siri AI, at its Worldwide Developers Conference. This updated version will launch later this year in beta form. Siri AI aims to improve the personal assistant experience by pulling context from various applications. For example, users can ask Siri about a podcast recommended by a friend via text, and Siri will provide the information without needing to navigate multiple apps. What's interesting is that this transformation includes a collaboration with Google. Apple secured an agreement to leverage Google’s Gemini AI models. However, Apple emphasizes that Siri AI is not simply a rebranded Google assistant. Craig Federighi, Apple’s senior vice president of software engineering, clarified that Apple uses none of Google's models or infrastructure deployed to their customers. Apple’s vice president of AI, Amar Subramanya, explained they developed new Apple Frontier Models using proprietary data, refined with outputs from Google’s Gemini Frontier models. Apple also enhanced its Private Cloud Compute infrastructure with Google and Nvidia to handle Siri AI's demands. The bottom line is that this new Siri AI promises more intelligent and seamless interactions for users.