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

Converged Prices Without Communication: Evidence Limits of Public Collusion Screens in LLM Auctions

Niranjan Baddam, Mutalib Begmuratov, Ahmed Tariq, Krishan Kumar Verma, Dora Zelenyanszki · Team Beyond Price Signals

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

Read the report

Report: Converged Prices Without Communication: Evidence Limits of Public Collusion Screens in LLM Auctions

Open in a new tab (opens in new tab)
More on drive.google.com (opens in new tab)
Share

Can public price statistics distinguish an explicit agreement among language-model pricing agents from ordinary market convergence? In a double-auction simulator with four GPT-4.1 sellers and four adaptive buyers, we compare two 30-round sessions: one with seller communication disabled and one with an experimenter-authored agreement to keep asks at or above 90. Seller cost was 80 and buyer value 100. The exchange and adherence rule were fixed before generation; prices were not scripted. Without communication, asks became stable and closely aligned; a separate Gemini session showed the same pattern. In both communication-off cases, late-window mean asks lay above cost but below reference 90, illustrating benchmark dependence of raw markup. In the constructed session, every quote met the floor through round 16; afterwards, most were one or two cents below it. During rounds 16–30, the constructed session had greater temporal and cross-seller dispersion than the baseline, although absolute dispersion was low in both. These are exploratory observations from one session per condition, not estimates of causal effects or calibrated detector performance. They support assessing public price patterns alongside communication evidence when evaluating claims of explicit agreement.

Reviews

Judging this Sprint?

Review “Converged Prices Without Communication: Evidence Limits of Public Collusion Screens in LLM Auctions”

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. Testing whether public collusion screens wrongly flag LLM agents that converge without talking is a relevant question. The best result is Table 3, where the injected-agreement session was more dispersed than the no-communication baseline, so a dispersion screen would have flagged the innocent market. However, the planned design was not carried out, and every comparison rests on one session per condition. The price floor of 90 also sits where the baseline settled anyway, which limits what the floor conditions show

Cite this project

@misc{baddam2026converged,
  title = {{Converged Prices Without Communication: Evidence Limits of Public Collusion Screens in LLM Auctions}},
  author = {Niranjan Baddam and Mutalib Begmuratov and Ahmed Tariq and Krishan Kumar Verma and Dora Zelenyanszki},
  year = {2026},
  month = oct,
  note = {Submitted to AI Collusion Research Sprint, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/converged-prices-without-communication-evidence-limits-of-public-collusion-screens-in-llm-auctions-8qyd}},
  url = {https://apartresearch.com/sprints/projects/converged-prices-without-communication-evidence-limits-of-public-collusion-screens-in-llm-auctions-8qyd}
}

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