Apple is bringing third-party AI models directly into its operating systems. At its Worldwide Developers Conference on June 8, Apple will unveil iOS 27, iPadOS 27, and macOS 27, all supporting a new AI extension system. These extensions allow installed apps to access generative AI capabilities on demand through Apple Intelligence features such as Siri, Writing Tools, and Image Playground.

The extensions work for generating and editing text and images, as well as handling conversations. Each AI model can carry its own distinct voice, so you always know which service is responding. Built-in Siri would use one voice while third-party models would use separate ones, giving users clear audio cues about who is talking.

Third-party developers will need to update their apps to support these extensions, but adoption seems likely. Users could switch between models seamlessly—Claude for writing, Gemini for images, ChatGPT for Siri conversations. This approach signals Apple has essentially abandoned building a competing AI platform, instead integrating existing services directly into its ecosystem.