Skip to content
Apart Research
Apart Research
Contact us
Sprint projectFeb 2, 2026Montreal
1st place

LidaSim: Testing AI Policies With Persona-Based Simulations

Linh Le, David Williams-King, Arthur Colle · Team Lida Safety

Submitted to The Technical AI Governance Challenge. Sprint projects are early-stage work by participants, not Apart Research publications.

Read the report

Report: LidaSim: Testing AI Policies With Persona-Based Simulations

Open in a new tab (opens in new tab)
Code (opens in new tab)
Share

We simulate well-known figures in AI and politics with agents, scraping large amounts of data to get realistic simulations. Then, we test questions and proposed policies against these public figures, to see which policies are more likely to be broadly supported. We focus on policies related to compute governance and hardware verification. We present a multi-agent simulation framework called LidaSim that orchestrates AI-powered agents to deliberate, debate, and vote on complex topics. Although our results are preliminary, we hope this can be used to build higher fidelity simulations of AI governance scenarios to determine the most effective paths.

Reviews

Judging this Sprint?

Review this project

Your public critique appears on this page without your name. Your private critique is not published; only the Apart team reads it. If you agree below, we share your review with grantmaking.ai (opens in new tab) and the Transformative AI Fund so strong projects can be funded.

Not shown on this page.

Shown on this page, without your name.

Only the Apart team reads this, and funders if you agree below.

Share my name publicly on grantmaking.ai *
Share my private critique with funders *

How much would this matter for AI safety if it worked? How innovative is it? For scores of 4-5: is this actually new to the field, or replicating recent work?

Scoring guide
  1. 1Negligible. No clear problem addressed, or no meaningful novelty.
  2. 2Limited. Addresses a real problem but with a generic or well-trodden approach. Incremental at best.
  3. 3Moderate. Clear problem with a reasonable approach; some novelty in framing or method beyond routine application of existing tools.
  4. 4Significant. Important problem with an original approach, or identifies a neglected problem area. A valuable contribution others could build on.
  5. 5Exceptional. Tackles a critical AI safety problem with a genuinely novel approach, or opens a new research direction. Clear theory of change. You'd be excited to share this with researchers in the area.

How sound are methodology, implementation, and findings?

Scoring guide
  1. 1Seriously flawed. Methodology broken, results uninterpretable, or implementation doesn't work.
  2. 2Weak. Approach has significant gaps: missing validation, flawed experimental design, or incomplete implementation.
  3. 3Competent. Technically solid given the short duration. Methodology makes sense, results are interpretable, limitations acknowledged, work builds toward clear conclusions.
  4. 4Strong. Thorough methodology with convincing validation. Results clearly support conclusions. Immediately useful for future work.
  5. 5Exceptional. Ambitious scope executed rigorously. Surprising findings, novel methods, or unusually robust validation.

How clearly are work, findings, and impact potential communicated?

Scoring guide
  1. 1Incomprehensible. Cannot determine what the project is actually claiming or doing.
  2. 2Hard to follow. Key information buried, missing, or diluted by excessive length. Significant effort to extract main points.
  3. 3Clear enough. Can understand the problem, approach, and results without undue effort. Core content clearly present: problem, method, findings, limitations.
  4. 4Well presented. Easy to follow, well-structured, appropriate level of detail. Target audience would get it quickly.
  5. 5Exceptionally clear. A pleasure to read. Complex ideas made accessible. Could serve as a model for how to present this type of work.

  1. The team shows clear technical talent—building a complete multi-agent deliberation system with a polished UI in a hackathon timeframe is impressive. My main criticism is that the team did not persuasively argue for why (or under what conditions) it is valid to draw inferences about actual human responses from LLM simulations of those responses. It seems interesting to figure out what a reasonable methodology for doing so would look like.

  2. This is a really creative and well-executed project. LLM persona simulations to stress-test governance policies seem well worth explorting, and the write-up is clear and no-nonsense. I especially appreciate the detailed appendix.

    Two suggestions for strengthening the work:

    Try to synthesize the simulation results into higher-level insights. What do current simulation outcomes tell us about which policies would work? And are there generalizable insights into which properties make proposals workable?

    The key open question for me is validation: are the simulation outcomes predictive, do they tell us something about which policies may actually succeed? I'm not sure how to best do it but backtesting seems like it might be worth a shot and results here would make it so much stronger as a project.

Cite this project

@misc{le2026lidasim,
  title = {{LidaSim: Testing AI Policies With Persona-Based Simulations}},
  author = {Linh Le and David Williams-King and Arthur Colle},
  year = {2026},
  month = feb,
  note = {Submitted to The Technical AI Governance Challenge, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/lidasim-testing-ai-policies-with-personabased-simulations-o8x6}},
  url = {https://apartresearch.com/sprints/projects/lidasim-testing-ai-policies-with-personabased-simulations-o8x6}
}

More from The Technical AI Governance Challenge

All 47 entriesBrowse all projects

Build something like this at the next Sprint

AI Collusion Research Sprint · Oct 23 - 25, 2026