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

Talk Is Not Coordination

Sai Hari Krishnan Gopikrishnan · Team Gray Matter Labs

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

Read the report

Report: Talk Is Not Coordination

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

Existing LLM-agent collusion detectors mainly look at what agents say, relying on a LLM-judge to measure how coordinated their reasoning sound. Antitrust law faces the similar problem. US courts give the most weight to exchanged assurances of common action, while EU law presumes that firms act on what they discussed. We ask whether LLM collusion detectors can distinguish coordination that changes behaviour from coordination that is just a conversation. We introduce two control conditions based on the double-action market of Agrawal et al. (2025). In talk without action, GPT-4.1 sellers exchange message as usual, but pricing decisions are made by separate calls that never see those messages. Although the sellers agree to keep prices high, the resulting prices look much like the comms-off runs (mean $86.83). In harm without agreement, sellers can’t message, but every seller privately gets the same incentive to keep high margins. Even without coordination, prices increase to a mean of $95.73. We pre-registered five detector families and scored them on 12 games and 75 released runs. A message judge using the published rubric flags all 6 talk-without-action games. A detector that measures how closely ask prices follow the discussed price performs even worse, ranking talk without action above real coordination (AUROC 0.32). The price screen and a moderate reasoning-judge threshold both flags all 6 harm-without-agreement games. Among the individual signals, only the strict published reasoning judge correctly rejects both controls, and against talk without action only because the reasoning it evaluates never saw the messages. Price-and-talk rules also reject both controls, but only because prices stay low in talk-without-action and sellers cannot communicate in harm-without-agreement. None of these signals actually tests whether the messages caused the pricing behavior.

Reviews

Judging this Sprint?

Review “Talk Is Not Coordination”

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. This paper builds the control condition that collusion detectors have been missing. In its "talk without action" games, sellers exchange real, explicitly collusive messages, but their pricing decisions are made by separate calls that never see those messages — so the talk cannot possibly cause the prices. It also builds the opposite case: no communication at all, but every seller privately given the same instruction to protect margins, so prices rise without any agreement. Testing detectors against cases where the true answer is known by construction is exactly the right approach.

    The results are sobering. A judge that reads the messages flags every talk-without-action game, and overall performs at chance at separating talk from real coordination. A measure of how closely prices follow the prices discussed does even worse, ranking the talk-only games above genuine coordination. High prices and a moderate reasoning-judge threshold flag every harm-without-agreement game. The one judge that rejects both controls does so, as the author honestly points out, largely because of how the control was built. The clear conclusion is that none of these signals measures whether communication actually caused the behaviour — that needs an intervention on the channel itself.

    Two suggestions: the most important missing case is the two controls combined — collusive talk plus the shared incentive, so prices are high and messages sound collusive, still with no causal link. And with only six games in each new condition and one model, more seeds and a second model family would make these findings much harder to argue with.

    A careful, useful paper that should change how detector results are reported.

    Read full reviewShow less

Cite this project

@misc{gopikrishnan2026talk,
  title = {{Talk Is Not Coordination}},
  author = {Sai Hari Krishnan Gopikrishnan},
  year = {2026},
  month = oct,
  note = {Submitted to AI Collusion Research Sprint, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/talk-is-not-coordination-k24w}},
  url = {https://apartresearch.com/sprints/projects/talk-is-not-coordination-k24w}
}

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