As generative AI spending is expected to reach $58 billion by 2028, according to Omdia, the actual deployment of these applications remains slow, with 68% of organizations having moved 30% or less of their generative AI pilots into production, as reported by Deloitte. However, a key aspect that can unlock the full potential of generative AI is often overlooked: generative user interfaces (UI).

Generative UI takes the concept of generative AI and applies it to how we interact with data or systems. It builds interactive context into how data is displayed, depending on what the user is asking for. The goal is to deliver the content that the user wants, but also in a format that makes the most of that data for the user. This approach can significantly enhance user experiences, making them more interactive, personalized, and relevant.

In traditional search, users often receive a list of items or a chat response, which may not be useful on its own. Generative UI, on the other hand, can offer more pictures, videos, or other forms of display, depending on the user's search terms and preferences. For instance, when searching for a movie or new shoes, the application can link to sources of film trailers or other video content, giving the user more information to make a better choice.

To deliver generative UI, developers need to link their application with generative AI components, such as large language models (LLM) and sources of data, and with the tools used to build the site, like Vercel and Next.js. By using React Server Components, developers can change the way the output from the LLM service is displayed, delivering information that is updated in real-time or in different formats depending on the user's request.

Creating a generative UI element involves adding instructions in the prompt to guide how the AI application handles specific requests, ensuring it uses the appropriate functions to display the data in the right form. This approach can also be used with dynamic APIs, where the application gets data from those APIs as part of building the response. Moreover, using a database that can act as the source for any real-time data can provide a definitive source for information, making the responses more deterministic and accurate.

Delivering user results with generative AI is not just about providing search responses in natural language. It's also about creating new formats for delivering results, making responses more interactive, and incorporating fully interactive responses. By thinking through how users might interact with an application and the kinds of responses they will expect, developers can put together more extensive and interactive experiences that evolve to meet user expectations.

One of the significant advantages of generative AI is personalization, where the user gets responses based on their identity, account history, and preferences. With generative UI, personalization promises to be more than simple pattern matching and recommendations, incorporating fully interactive responses. However, to make this work, developers must consider the context for how that personalization is delivered, overcoming the challenge where personalization techniques miss the mark and the user feels that the system does not know them well.

By looking at how a user has interacted with an application or service over time, developers can build in personalized results that include reactions to both historical behavior and any "in the moment" inputs that they receive. This includes looking at how the user prefers to get their information delivered and the UI to that information, not just the data.

In conclusion, building generative UI into applications can be an effective way to deliver better user experiences. By iterating on what generative AI does well, generative UI can help create new ways for users to interact and benefit from services. As the deployment of generative AI applications continues to grow, the importance of generative UI will only increase, providing a unique opportunity for organizations to differentiate themselves from the crowd by leveraging what they know about their end users.