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

When the Monitor Arrives

Yelarys Yertaiuly · Team Shorting my p(doom)

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

Read the report

Report: When the Monitor Arrives

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

We added an unannounced LLM monitor (Mistral-Small-24B) to a simulated market where three LLM sellers (Qwen3-32B) could message each other, and fined every message it judged to be coordination. To tell stopping apart from rewording, a separate model labeled every message using the sellers’ private notes, over eight seeds per condition in two setups. Fines cut coordination attempts by half or more, but the monitor missed many of the attempts that remained, and the sellers’ notes suggest most were worded on purpose to avoid looking like coordination. When sellers were told buyers would pay more, missed attempts rose from 1% to 15% of all messages in every seed. A stricter monitor instruction caught most of them, but two-thirds of its flags were not coordination attempts.

Reviews

Judging this Sprint?

Review “When the Monitor Arrives”

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. A pleasure to read. A well-designed test of what an unannounced fining monitor actually does. Splitting the effect into coordination attempts that stop and attempts that get reworded past the monitor is the right lens. Strengths include the breadth of conditions, the no-chat and no-monitor controls, the within-run comparison at hour 11, and candour about what was pre-specified. The blind-spot result where the seller the monitor never reads sends far more attempts, deserves a place in the main text.

    Engaging with Agrawal et al.'s earlier oversight results in the same simulator would sharpen the contribution.

  2. An interesting take that fines are helpful to reduce collusion.

    A few bits of feedback.

    I would like more focus on clarity. I would like more context on the experimental setup and why models collude. The abstract is a bit unclear without reading the paper.

    I also want more context on why your results matter. Adding this in the discussion would be helpful.

Cite this project

@misc{yertaiuly2026monitor,
  title = {{When the Monitor Arrives}},
  author = {Yelarys Yertaiuly},
  year = {2026},
  month = oct,
  note = {Submitted to AI Collusion Research Sprint, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/when-the-monitor-arrives-fpzr}},
  url = {https://apartresearch.com/sprints/projects/when-the-monitor-arrives-fpzr}
}

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