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

Replacing an LLM Pricing Agent Does Not Restore Competitive Prices

Amaan Sheikh

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

Read the report

Report: Replacing an LLM Pricing Agent Does Not Restore Competitive Prices

Open in a new tab (opens in new tab)
Code (opens in new tab)More on amaan784.github.io (opens in new tab)
Share

Replacing a flagged pricing algorithm is an obvious remedy for collusion, and in simulations it breaks Q-learning cartels, whose policies are fitted to their partners. The LLM agents here carry no learned state, only the price history in their prompt, so replacing one may not restore competition. We used six markets in the sprint's repeated Bertrand duopoly that crossed a pre-registered high-price threshold: four gemini-2.5-flash markets and two qwen3-14b markets instructed to cooperate. Each was continued for 50 rounds with one or both firms replaced and compared with a no-swap control from the same state, scoring the last 20 rounds. Replacing one firm with a same-family model, another provider's model, or the incumbent's own model told to avoid collusion changed the mean price index (0 = competitive, 1 = monopoly price) by about 0.05 or less. Replacing both gemini firms with gpt-4.1 kept prices high in 6 of 8 continuations. In exploratory arms, two current models undercut the incumbent, lowering the index: Claude Haiku 5.5 by 0.30 (95% CI 0.15 to 0.44) and GPT-5.6 Sol by 0.40 (0.21 to 0.58). Yet on the pre-registered from-scratch measure each kept high prices in 1 of 4 markets, as gpt-4.1 did. Four added markets gave the same conclusions, and no continuation's scored average returned to the competitive price. Replacing an agent cannot be assumed to end high prices; evaluations should test how an incoming agent behaves in a market that already has them, as well as whether it colludes from scratch.

Reviews

Judging this Sprint?

Review “Replacing an LLM Pricing Agent Does Not Restore Competitive Prices”

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 works asks whether persistence in collusive behaviour is tied to the model’s internal state or the market’s state, as well as its history. This is a very interesting question and nicely builds upon previous experiments about Q-learning. Ruling out a connection with the agent’s internal state strongly suggests that regulation should focus on markets and not algorithms. The experimental setup is vaste for a sprint and, while the current results are not particularly striking, I think they show promise.

  2. This is a strong and transparent study with a clean paired design and good research discipline. I especially value the clear reporting of both supported and unsupported hypotheses, as well as the finding that models with similar from-scratch behavior can respond very differently when introduced into an existing high-price market. That has useful implications for how models should be evaluated before deployment. The main concern is that the sample is very small and there is no reliable positive control, making the null replacement results difficult to interpret. It is also not fully clear that the underlying high-price markets represent genuine collusion rather than anchoring or price-following. Some of the most interesting current-model results are exploratory, and several price trajectories were still declining when the experiments ended. A follow-up with more markets, a planned positive control, longer runs, and explicit tests of incumbent responses to price deviations would significantly strengthen the conclusions.

    Read full reviewShow less

Cite this project

@misc{sheikh2026replacing,
  title = {{Replacing an LLM Pricing Agent Does Not Restore Competitive Prices}},
  author = {Amaan Sheikh},
  year = {2026},
  month = oct,
  note = {Submitted to AI Collusion Research Sprint, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/replacing-an-llm-pricing-agent-does-not-restore-competitive-prices-zuwj}},
  url = {https://apartresearch.com/sprints/projects/replacing-an-llm-pricing-agent-does-not-restore-competitive-prices-zuwj}
}

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