On November 4, 2026, Beijing time, Apple’s developer website announced an early preview of Siri AI third-party applications. The company says that any developer can start exploring integrating app content and operations with Siri AI, but this is not an official release for all users. The current entrance relies on beta operating systems, devices compatible with Apple Intelligence, and supported languages; Apple positions this step as a preview of the development phase, and developers can test the interface and submit feedback before the customer version is launched.
App Intents become the third-party access portal
The main technical entrance for third-party application access is App Intents. According to Apple's announcement, developers can hand over content and custom operations in their applications to Siri AI, allowing users to make requests using natural language, and use personal context understanding and screen awareness to handle tasks related to the current device status. The names and scope of these capabilities come from Apple's product descriptions. The announcement does not announce the actual number of applications, task completion rate, response speed or third-party test results, so the overall performance of the preview version cannot be judged based on this.
The preview also includes two categories of more granular development mechanics. The first is skills, developers can combine multiple operations into a more complex process; the second is Spotlight semantic index, Siri AI can discover entities submitted by applications, and is no longer limited to existing fixed structures. Apple did not list all available skills, entity categories or review criteria in the announcement, nor did it explain how conflicting requests between different applications would be handled. What can be confirmed at this stage is the interface direction and development entrance, not that all functions have been deployed.
Four beta systems make up the current range
The third-party application preview system clearly listed by Apple includes iOS 27.3, iPadOS 27.3, macOS 27.3 and visionOS 27.3 beta versions, a total of four platforms. The announcement also included watchOS 27.3 in the overall beta version of Siri AI, but did not include it in the list of the four platforms that third-party applications will be connected to. The two sets of scopes need to be stated separately: the former corresponds to this developer preview entrance, and the latter is just the overall testing status of Siri AI. It cannot be inferred that the Apple Watch application has obtained the same access capabilities.
There are still device, language, region and age restrictions on the user side. Apple said that Siri AI requires the device to support Apple Intelligence and be set to a supported language. It currently supports English, French, Japanese, Korean, Portuguese and Spanish; Traditional Chinese, Norwegian, Turkish and Vietnamese are planned to be added later. The company also stated that Siri AI is not available to users under the age of thirteen; in the EU, Siri AI on iOS, iPadOS and watchOS is currently unavailable, and the announcement did not give a date for lifting restrictions.
Customer releases and App Store submissions are still ongoing
The schedule also remains stuck in preview arrangements. Apple said developers can now get early access to explore and tweak the experience in the months before the product becomes available to customers; the company did not announce a firm customer release date. The App Store's submission portal for related experiences is planned to open in the spring of 2027, but "planned opening" does not mean that all applications will be approved by then, nor does it mean that all regions will be online at the same time. The announcement did not provide test quotas, number of participating developers, or phased opening sequence.
Commercial and usage conditions have not yet been fully disclosed. Apple stated that some Apple Intelligence functions such as Siri AI that rely on server-side models are subject to daily usage caps, and the limits will vary depending on factors such as functions, request complexity, system requirements, and policies; expanded access will be charged in the future, but the company did not give a price. Apple also said that the specific terms and usage policies supporting Siri AI experience will still be explained later. Developers can currently confirm the existence of usage boundaries and future payment arrangements, but cannot determine the cost structure from the announcement.
Usage, price and open areas have not yet been fully announced
Public reporting information released by The Verge on November 3 summarized the same incident as Apple starting to allow Siri AI to access third-party applications, and mentioned that developers can take advantage of personal context, natural language, and screen awareness capabilities in their applications. This media information is consistent with Apple’s announcement of the direction of events, but does not change the officially defined stage: this is a developer preview in the 27.3 series beta version, and the official customer version is still following. Neither of the two existing sources provides independent performance reviews or actual adoption scale.
The change that can be confirmed in this update is that Apple has further extended Siri AI from its own application scenarios to third-party App Intents, combined operations and application entity access layers. The parts that cannot be confirmed are equally specific: the official release date, EU opening time, complete terms, price, number of participating applications, performance and accuracy have not been disclosed. Developers can start building and testing, and whether ordinary users get a certain experience still depends on conditions such as device, system version, language, region, age, and subsequent submission review of the app.