This project emerged from our experiments with custom AI assistants in a 2023-2024 AI Faculty Working Group. We noticed how quickly people generated new ideas after seeing just a few concrete examples. What proved particularly powerful was how one use case would spark recognition of entirely different possibilities. For instance, someone might initially think of AI primarily as an information source, but seeing examples of AI as a Process Coach or Reflective Guide would reveal its potential for guiding experiences and supporting complex processes rather than simply providing answers.
As we developed more custom chatbots, a few patterns seemed to emerge. We found that having concrete examples helped others envision possibilities more quickly and effectively.
This collection of AI roles and examples aims to bridge the gap between abstract capabilities and practical applications by combining pattern descriptions with real examples of AI assistants and prompts.
The goal is to help people:
The collection is very much a work in progress and given how fast things are moving and how many possible uses there are for GenAI, it might be impossible to maintain and of limited use. But for now, it’s a provisional experiment in iterative meaning-making at time when we’re all just trying to find our way.
This collection is meant to be a kind of pattern library, a way of sharing solutions to common problems in a way that people can take them and adapt them to their own contexts.
In this case, we’re trying to map out some emerging ways of using AI to solve problems in higher education as we learn work with a design material — generative AI.
These patterns serve multiple purposes: