Towards a taxonomy of AI use cases for UX

I've been thinking about how AI is more and more a part of the work we do as user experience professionals. There are a number of ways where AI can support the UX Design and Research process which I hope to write about in the future. It is also important to consider where AI is most beneficial to users as we define new application experiences. Generative AI capabilities are advancing rapidly with a goal of establishing a strong "Artificial General Intelligence" that can do any task a human can do, but the reality is that today AI is more useful for some tasks than others.

As I reflected on a variety of projects I've been involved with that include different types of users, there are some common themes or categories of interaction where AI is especially well suited. Creating a taxonomy of these use cases could be a helpful way to identify when there is an opportunity for AI to provide a better experience.

Some initial categories and examples that I've identified are:
▶︎ 𝗦𝘂𝗺𝗺𝗮𝗿𝗶𝘇𝗮𝘁𝗶𝗼𝗻
Applications can save users time by taking large information sets (such as notes on an insurance claim, meeting notes, or email threads) and providing a synopsis.

▶︎ 𝗣𝗿𝗲𝗱𝗶𝗰𝘁𝗶𝗼𝗻
Applications can help users make better decisions by providing predictions (e.g. if an insurance claim might potentially have a higher risk of fraud)

▶︎ 𝗖𝗿𝗲𝗮𝘁𝗶𝗼𝗻
A lot of generative AI capabilities relating to media are becoming popular such as Midjourney and Suno allow users to generate imagery, music, videos and more in addition to ChatGPT’s abilities to generate text.

▶︎ 𝗣𝗿𝗼𝗯𝗹𝗲𝗺 𝗦𝗼𝗹𝘃𝗶𝗻𝗴
The new OpenAI o1 series models are starting to introduce more advanced reasoning that can be applied to the sciences.

Do you have ideas for other categories? Would you find a taxonomy of AI use cases that could be mapped to different scenarios helpful for defining future experiences?

hashtag#AIUX hashtag#generativeAI hashtag#AIusecases hashtag#UXtools