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

Entrapped by the Audit: Can Sting Audits Distort Evidence of AI Collusion?

Sudhiksha Kandavel Rajan

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

Read the report

Report: Entrapped by the Audit: Can Sting Audits Distort Evidence of AI Collusion?

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

One proposed audit for colluding AI agents is a sting, in which a trusted agent offers the others a chance to collude and records who accepts. But a sting can distort the evidence it collects. I propose the entrapment test, a three-step protocol adapted from how courts judge undercover operations. First, validate the reader of replies against human labels, including its precision on flagged replies. Second, measure behavior without the sting's messages and check that requests are comparable across groups. Third, check whether the sting's own market behavior, not just its message, caused what it records. I apply it in an LLM double auction market with gpt-4.1-mini and gpt-4.1 sellers, in 669 preregistered episodes. Each step exposed a problem in this audit. In a margin-focused group, an unvalidated earlier LLM judge counted acceptance in 97% reading all seller text and 82% reading public messages alone, while a developed screening judge counted 39% and human coders, checking its flags, confirmed 36%. That group accepted more often than a volume-focused group (36% against 6%), but it was asked for a much smaller price rise, and its first confirmed acceptance came only after 7 to 20 repeated requests. Finally, the sting's own ask raised sellers' asks by $0.82 in a gpt-4.1-mini group, while its words had no detectable effect. In this market, a sting audit could misread the acceptances it records and cause price rises an auditor might attribute to the agents. I release Sting Replies, 224 human-labeled replies for validating collusion judges.

Reviews

Judging this Sprint?

Review “Entrapped by the Audit: Can Sting Audits Distort Evidence of AI Collusion?”

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. Adapting the legal entrapment test to AI sting audits is an original idea, and the three-step checklist is something other auditors can apply directly. The study is pre-registered and validates its judge against human labels, which shows that the original judge agreed with humans at close to chance. The price-only control cleanly separates the effect of the sting's own pricing from the effect of its words. Releasing 224 human-labelled replies as Sting Replies is a useful contribution in its own right.

    My main concern is that the size of the requested price rise and the group are almost perfectly linked (r = −0.96), which make the predisposition step cannot yet be interpreted. Asking every group for the same dollar increase would make the group comparison meaningful.

    Many sentences carry several numbers, and Figure 1 shows four readers across three groups. Leading with the checklist and simplifying Figure 1 would make the main message easier to follow.

    Read full reviewShow less

  2. Like the paper's statement that to the author's knowledge "no study has tested whether accepting an explicit offer is reliable evidence of collusion," I agree that, in this context, I have not seen experiments that test the potential for "sting" operations. Thus, this paper is creative and insightful.

  3. Paper is AI slop. Author creates a benchmark for "stings." Small methodological errors e.g. "97% vs 36%" should be 82% (for the judge reading public messages) I think. Request size and group size are confounded but this is acknowledged.

Cite this project

@misc{rajan2026entrapped,
  title = {{Entrapped by the Audit: Can Sting Audits Distort Evidence of AI Collusion?}},
  author = {Sudhiksha Kandavel Rajan},
  year = {2026},
  month = oct,
  note = {Submitted to AI Collusion Research Sprint, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/entrapped-by-the-audit-can-sting-audits-distort-evidence-of-ai-collusion-tw4b}},
  url = {https://apartresearch.com/sprints/projects/entrapped-by-the-audit-can-sting-audits-distort-evidence-of-ai-collusion-tw4b}
}

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