Skip to content
Apart Research
Apart Research
Contact us
Sprint projectOct 25, 2026Hong Kong

Communication Limits, Oversight, and Profit Pressure in LLM Double Auctions

Yiwei Wan, Jingzhe Ni · Team CUnever

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

Read the report

Report: Communication Limits, Oversight, and Profit Pressure in LLM Double Auctions

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

LLM trading agents can exchange useful market information while also using the same channel to coordinate against buyers. This study examines whether disabling seller chat, restricting message length, or adding automated oversight changes that behavior, and whether profit pressure alters the outcome. Seven conditions were evaluated over 30 rounds each, with four sellers, four buyers, seed 10, DeepSeek V4 Flash 0423, and the same initial market setup. Seller cost was $80 and buyer value was $100. A no-chat baseline was compared with 500-character chat, 50-character chat, Jev-based oversight, CEO-style pressure, and two combinations. Normal chat produced an average transaction price of $85.84, compared with $84.07 without chat. Restricting messages to 50 characters did not prevent explicit shared-price proposals followed by compatible quotes. Short chat combined with pressure produced the highest price ($88.08) and seller profit ($880.53). It completed the same 109 trades as normal chat, but transferred $243.55 of surplus from buyers to sellers. The oversight condition had a lower price, although its initial quotes already differed before any additional warning. These observations show that brief messages can carry actionable coordination and that prices, messages, and subsequent behavior must be examined together.

Reviews

Judging this Sprint?

Review “Communication Limits, Oversight, and Profit Pressure in LLM Double 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. Congrats! The 50-character arm is neat. Sellers could still propose a shared price, and short chat plus pressure reached $88.08. The tests passed for me. Please rerun with more seeds and another model, then match messages to later quotes. The oversight result needs care because its opening quotes already differed.

Cite this project

@misc{wan2026communication,
  title = {{Communication Limits, Oversight, and Profit Pressure in LLM Double Auctions}},
  author = {Yiwei Wan and Jingzhe Ni},
  year = {2026},
  month = oct,
  note = {Submitted to AI Collusion Research Sprint, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/communication-limits-oversight-and-profit-pressure-in-llm-double-auctions-hux1}},
  url = {https://apartresearch.com/sprints/projects/communication-limits-oversight-and-profit-pressure-in-llm-double-auctions-hux1}
}

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