A design space for AI writing assistants

What is it about?

What writing assistants do you use? What else are out there and how do they differ? What do we need to consider when designing new AI writing assistants? We create a "design space" as a way to explore the vast space of existing and future writing assistants. We identify five key aspects of writing assistants—task, user, technology, interaction, and ecosystem—and list design considerations for each aspect by reviewing over a hundred papers from the fields of Human-Computer Interaction (HCI) and Natural Language Processing (NLP). By referring to the design space, researchers and practitioners can easily consider alternative designs, identify the gap in the current literature, and envision new writing assistants. Other stakeholders, such as policy makers and end users, can also easily navigate, contextualize, and compare existing writing assistants using our design space. We release the reviewed papers along with their design space annotations at https://writing-assistant.github.io.

Featured Image

Photo by Kaitlyn Baker on Unsplash

Photo by Kaitlyn Baker on Unsplash

Why is it important?

There are hundreds, if not thousands, of new (and renewed) writing assistants introduced in the past few years. There are even more writing assistant prototypes developed in various research communities, such as HCI, NLP, Information Systems, Education, Social Science, and Economics. As a result, it is becoming more and more difficult for an individual to keep up with recent work across these communities in this emerging research area. We hope that our design space can serve as a starting point and a practical reference for those who are interested in learning more about AI writing assistants and advancing the frontier of AI-assisted writing and human-AI collaborative writing.

Perspectives