AI could make institutions better or worse at reasoning and coordinating. Prototype one tool or evaluation from BlueDot’s decision-making hackathon or Forethought’s design sketches.
Better collective decisions are one of the few things that help with almost every AI risk at once.
Output: A prototype or evaluation with a short demo.
Read the list of ideas above and pick the one you would most enjoy.
Check the source to see whether someone has already done it.
Scope it down to something you can finish in your available time.
Do a quick first version and show it to someone.
Write up what you made and share it with the source authors.
All steps done
Share what you made, then try another project. The PDF version has room for notes.
Lange argues AI outputs should count as one input alongside our own reasons, not replace them. Run a small study: show volunteers AI answers (some reliable, some subtly wrong) and measure how much they update.
Pick one decision-maker, one AI risk and one decision they actually face. Lay out the options and make a clear recommendation in two pages.
Gather 6–12 friends, give each a stakeholder role (lab, regulator, journalist, public) and play out how society responds to a fictional AI incident.
