AI assistants are reshaping design around context across apps
Designers are shifting from building isolated app tasks toward shaping how context moves between products, with Apple's App Intents framework offered as an example of how AI assistants can carry a user's intention across applications.
Designers are beginning to treat the space between apps as the main challenge of digital experience, as AI assistants take on the job of carrying a person's intention from one product to another. Rather than asking users to find the right application and navigate its features, the emerging model starts with what the person wants to accomplish and works out which information and capabilities are needed to get there.
The article illustrates the problem with a common scenario. A friend recommends a restaurant over a message, and the user must check the location, look at a calendar for availability, return to the conversation, and write a reply. Each step is simple on its own, but together they fragment the experience, forcing the person to remember what they were doing and what comes next as they move between products. The author argues that this is a familiar pattern, since products have long been designed around individual tasks and have left people to connect those tasks themselves.
Apple's work on Siri and Apple Intelligence is presented as one example of the new direction. Through its App Intents framework, developers describe actions and content that can be made available to supported system experiences, giving Siri and other features a structured way to discover what an application can do. Under that approach, a user could ask Siri to find a restaurant mentioned in an earlier conversation, check availability for Friday, and help prepare a reservation. The applications remain responsible for their own data and functionality, but their capabilities become reachable from an experience that begins elsewhere, depending on the integrations and permissions available.
This shift changes the design problem itself. The challenge is no longer limited to helping someone navigate an application, but extends to making the right information and capabilities available in the right context, at the moment they become useful. The article describes context as a design material: it includes the current conversation, a person's preferences, previous decisions, the state of an ongoing task, and constraints such as whether a calendar is accessible or whether an action requires permission. A request as simple as finding dinner nearby may lack a budget, cuisine, or time, so the system must decide whether it has enough information to proceed or should ask a follow-up question.
Memory raises its own design questions. An assistant might remember that a friend recommended a particular restaurant or that Friday is the preferred evening, but the article stresses that remembering information is not the same as knowing it remains relevant. Systems should surface the context they intend to use, particularly when it shapes a decision, so that a person can confirm or correct it before the assistant acts. Showing that a recommendation came from a friend months earlier gives the user a chance to update outdated assumptions, keeping continuity from turning past behavior into an unquestioned premise.
The author also points to the role of product structure in making this work. Within the App Intents framework, developers define entities, parameters, and actions so that a system can understand what an application contains and how its capabilities can be used. These definitions influence whether the system can identify the correct item, interpret a request, and carry out an action reliably. In this view, the choices developers make about how information is structured shape the interaction as much as the visible interface does, determining how well an assistant can move between products on a person's behalf.
6
