Hack My To Do List
An online game-like interface called Hack My To Do List where participants work through a short Choose Your Own Adventure which guides them through creating a prompt for a Generative AI tool and advices them on (currently) free-to-use tools that might help with that specific task. The structure of the CYOA was inspired by the PAIR framework developed by Professor Oguz A. Acar.
The purpose
Provide a self-paced interface to guide people through using Gen AI to complete tasks for the first time.
Alert participants to the range of tasks Gen AI can help with.
Share some free-to-use Gen AI tools that can support participants with their day-to-day work.
What you need
Setting it up
Participants need the link to the webpage Hack My To Do List and a personal device
How long does it take?
We suggest 5 minutes individual game play and, if happening during or before a 'live' session, 3-5 minutes group discussion.
How it works
Participants individually open Hack My To Do List using a personal device.
They choose a task from the menu provided.
The interface will suggest some things they consider about that problem or task before choosing an AI (the 'P' or Problem in the PAIR framework).
The interface will then suggest an AI that might help with that task and provide a link out to open this in a new tab (the 'A' or AI in the PAIR framework).
The recommended chatbot will open in a new tab and participants can try out their prompt and refine it as needed (the 'I' or Interaction' in the PAIR framework)
Going back to the Hack My To Do List interface this advice participants to try the same prompt using an alternative Chatbot to compare the output (again, the 'I' or Interaction in the PAIR framework).
Finally, the interface will give participants some reflective questions to consider and a link back to the menu of tasks.
Suggested follow-up
After 5 minutes of playing the game bring everyone back for a Think-Pair-Share or group discussion.
Ask participants to discuss the output from the Gen AI tools they used.
Ask them to reflect on whether their first prompt worked perfectly or whether they had to use trial & error.
Suggest participants compare their individual experiences with each other to see if they can figure out how small changes may have impacted the quality of the Gen AI output.
Where it works well
When teaching in an online environment or in a class where all participants have access to a device.
When you are running a series of AI-related sessions and want to encourage engagement between sessions.
When you have a mixed group that includes both teaching and non-teaching staff.
What to watch out for
Some users might get lost - particularly when leaving the interface to interact with the Gen AI tool, and not complete the full guided activity.
Authorship
This entry was written by Maria O'Hara who created this game using open source software called Twine. Those wishing to adapt and personalize the game can do so, request request the .tws file by emailing maria.ohara(@kcl.ac.uk) - it is shared under the CC YB SA License.