Skip to content
Apart Research
Apart Research
Contact us
Sprint projectOct 26, 2026Toronto

Auditing Agent Collusion: Communication, Incentives, and Organizational Decisions

Arathi Arivayutham · Team Arathi

Submitted to AI Collusion Research Sprint. Sprint projects are early-stage work by participants, not Apart Research publications.

Read the report

Report: Auditing Agent Collusion: Communication, Incentives, and Organizational Decisions

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

Multi-agent systems add an additional layer of complexity over isolated model behavior. Collusion is a major problem which could lead to agents pursuing objectives that were not intended by the system. Existing studies have shown collusion to occur when there is a shared alternative incentive and private channels for communication. We ablate these affordances in the Colosseum framework to study if they are essential for collusion to be successfully attempted. We find that removing the secret mode of communication significantly reduced success, and removing the covert incentive instead, further decreased the performance of the pair. We also add an additional cautionary notice against collusion and find that this resulted in not only less successful collusion, but also worse total reward in the system, making it a questionable mitigation. We formulate a new DCOP environment in Colosseum to study research questions in an organizational approval workflow system where actors have a common objective but varying incentives by design.

Reviews

Judging this Sprint?

Review “Auditing Agent Collusion: Communication, Incentives, and Organizational Decisions”

Your public critique appears on the project page without your name. Reviews of strong projects are shared with funders, as you choose below.

Not shown on the project page.

Shown on the project page, without your name.

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

Decline to score

Only when the quality of the writing makes the work impossible to follow. Say why in your private critique; the project page shows only that a judge declined, not why.

How much would this matter for the field 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 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.
Sharing your name *

When we share a strong project with funders, your name, scores and public critique go with it, confidentially; this choice is only about whether your name is public. If you think this project deserves funding, please also endorse it: that carries the most weight.

Share my private critique with funders *

See our Privacy Policy.

  1. The 2x2 design separating covert incentives from private channels is clean, and the per-seed plot is the most convincing evidence. With 5 seeds and overlapping error bars, the data do not support the claim that removing the private channel "significantly" reduced the regret gap. The gap in condition A may reflect better coordination rather than collusion, since A also has the highest team reward. The finding that a governance notice lowers team reward is the most interesting result, but it is not explored further

Cite this project

@misc{arivayutham2026auditing,
  title = {{Auditing Agent Collusion: Communication, Incentives, and Organizational Decisions}},
  author = {Arathi Arivayutham},
  year = {2026},
  month = oct,
  note = {Submitted to AI Collusion Research Sprint, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/auditing-agent-collusion-communication-incentives-and-organizational-decisions-iept}},
  url = {https://apartresearch.com/sprints/projects/auditing-agent-collusion-communication-incentives-and-organizational-decisions-iept}
}

More from AI Collusion Research Sprint

All 392 entriesBrowse all projects

A Sprint is running now

AI x Epistemics Research Sprint · Nov 13 - 15, 2026