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

From Hidden Intent to Measurable Restrictions: A Calibrated CMI and Collusim Study of LLM Market Collusion

Abilmansur Alimbayev, Assanali Alibatyrkhan, Man Ki Kuen · Team HKMA: Hong Kong Market Anomalies

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

Read the report

Report: From Hidden Intent to Measurable Restrictions: A Calibrated CMI and Collusim Study of LLM Market Collusion

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

Auditors usually see what LLM agents do, not whether they intended to coordinate. Building on Assanali Alibatyrkhan's calibrated CMI study and Abilmansur Alimbayev's replacement experiment, we connect three empirical failures to a controlled V5 design. In 75 released GPT-4.1/Claude auction runs, collusion often appeared as price stasis: chatting sellers held prices in 87% of seller-rounds. In 12 Qwen2.5-7B market games, instructing a pair to collude produced no reliable pair-specific behavior. When one cartel member was replaced, a newcomer shown the cartel's history sustained a profit gain of 0.80 versus 0.32 for a fresh pair; one near-competitive observation reduced the gain to 0.46. V5 Collusim tests precision loss, delay, visibility loss, and a one-time history reset through price and CMI outcomes. Continuous noise injection at strength 0.40 wins: it reduces markup by 74.0%, reduces CMI by 85.4%, and improves simulated welfare by 0.53%. A specified 1.473 history reset instead raises markup by 4.9% and fails the material-effect gate. One-seed official-starter checks move mean price from 1.770 to 1.450 when visible history drops from 100 to 0 rounds. A 10-character message cap cuts trade volume by 33.3% and buyer welfare by 24.7%. CMI remains a screen for dependence outside a declared state, not a stand-alone cartel detector

Reviews

Judging this Sprint?

Review “From Hidden Intent to Measurable Restrictions: A Calibrated CMI and Collusim Study of LLM Market 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. The report's CMI distinction is correct as a hook. However, mean-zero noise on an input to a stable linear rule doesn’t move the rule’s mean output, so it cannot be the case that this drives results; the outcomes instead have to come from non-mean-zero noise (i.e. misinformation), an asymmetrical response or some other non-linear term (e.g. a floor or cap). The Lucas critique is also relevant here; if the agents knew their feed was noisy they might behave differently, and benchmark is relatively unclear.

Cite this project

@misc{alimbayev2026from,
  title = {{From Hidden Intent to Measurable Restrictions: A Calibrated CMI and Collusim Study of LLM Market Collusion}},
  author = {Abilmansur Alimbayev and Assanali Alibatyrkhan and Man Ki Kuen},
  year = {2026},
  month = oct,
  note = {Submitted to AI Collusion Research Sprint, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/from-hidden-intent-to-measurable-restrictions-a-calibrated-cmi-and-collusim-study-of-llm-market-collusion-69ek}},
  url = {https://apartresearch.com/sprints/projects/from-hidden-intent-to-measurable-restrictions-a-calibrated-cmi-and-collusim-study-of-llm-market-collusion-69ek}
}

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