AI Safety Sandbox
Search
PolicyGovernanceA weekendStarterFree

Run an AI policy tabletop exercise

Gather 6–12 friends, give each a stakeholder role (lab, regulator, journalist, public) and play out how society responds to a fictional AI incident.

Download PDFPeopleReading & writing

Why it matters

Tabletop exercises surface blind spots and objections that documents miss. They also work well as a society event.

Output: A scenario pack you can reuse, plus a summary of lessons.

Steps

0 of 5 done

  1. Write the scenario

    2 hr

    Invent a plausible AI incident and three “injects” (new developments you reveal during the game). Keep it fictional and non-technical.

  2. Write role cards

    1 hr

    One card per stakeholder with goals, constraints and secret information.

  3. Run it

    2 hr

    Play in rounds: reveal an inject, let teams discuss, then each announces a decision. Keep time strictly.

  4. Debrief

    30 min

    Ask: what surprised you? Where did coordination fail? What rule would have helped?

  5. Write up the lessons

    1 hr

    Summarise the decisions and lessons, and share the scenario pack so others can run it.

Related projects

Governance1–3 hoursGuided

Host a one-hour AI safety discussion

Pick one short reading, invite 4–8 people, and run a focused one-hour discussion with three good questions.

PeopleView project
PolicyA few weeksIntermediate

Build an AI tool that helps people make better decisions

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.

CodeVisualView project
PhilosophyA few weeksStarter

Run your own collective constitution

In Anthropic and CIP’s study, about 1,000 Americans used the open-source tool Polis to write a constitution for an AI. Run the same process with 30–100 students and compare your result with theirs.

PeopleReading & writingView project

Further reading

Related mentored projects

All mentored projects