# Breaking Algorithmic Collusion in Human-AI Ecosystems

arXiv: https://arxiv.org/abs/2511.21935 · HTML: https://mlsanity.com/paper/2511.21935/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 33 references that are in the map.
- First posted on arXiv 26 Nov 2025.

## Builds on (most cited first, 13 of 13)

- [Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm](https://mlsanity.com/paper/1712.01815/) (2017)
- [On the Utility of Learning about Humans for Human-AI Coordination](https://mlsanity.com/paper/1910.05789/) (2019)
- [Secret Collusion among AI Agents: Multi-Agent Deception via Steganography](https://mlsanity.com/paper/2402.07510/) (2024)
- [LLM as a Mastermind: A Survey of Strategic Reasoning with Large Language Models](https://mlsanity.com/paper/2404.01230/) (2024)
- [Algorithmic Collusion by Large Language Models](https://mlsanity.com/paper/2404.00806/) (2024)
- [Econometrics for Learning Agents](https://mlsanity.com/paper/1505.00720/) (2015)
- [Auctions Between Regret-Minimizing Agents](https://mlsanity.com/paper/2110.11855/) (2021)
- [Convergence Analysis of No-Regret Bidding Algorithms in Repeated Auctions](https://mlsanity.com/paper/2009.06136/) (2020)
- [Correlated and Coarse equilibria of Single-item auctions](https://mlsanity.com/paper/1601.07702/) (2016)
- [Nash Convergence of Mean-Based Learning Algorithms in First-Price Auctions](https://mlsanity.com/paper/2110.03906/) (2021)
- [Regulation of Algorithmic Collusion, Refined: Testing Pessimistic Calibrated Regret](https://mlsanity.com/paper/2501.09740/) (2025)
- [Algorithmic Collusion Without Threats](https://mlsanity.com/paper/2409.03956/) (2024)
- [Homogeneous Algorithms Can Reduce Competition in Personalized Pricing](https://mlsanity.com/paper/2503.15634/) (2025)

## Built on by (most cited first, 1 of 1)

- [Revisiting the Bertrand Paradox via Equilibrium Analysis of No-regret Learners](https://mlsanity.com/paper/2602.21620/) (2026)

## Abstract (arXiv)

AI agents are increasingly deployed in ecosystems where they repeatedly interact not only with each other but also with humans. In this work, we study these human-AI ecosystems from a theoretical perspective, focusing on the classical framework of repeated pricing games. In our stylized model, the AI agents play equilibrium strategies, and one or more humans manually perform the pricing task instead of adopting an AI agent, thereby defecting to a no-regret strategy. Motivated by how populations of AI agents can sustain supracompetitive prices, we investigate whether high prices persist under such defections. Our main finding is that even a single human defection can destabilize collusion and drive down prices, and multiple defections push prices even closer to competitive levels. We further show how the nature of collusion changes under defection-aware AI agents. Taken together, our results characterize when algorithmic collusion is fragile--and when it persists--in mixed ecosystems of AI agents and humans.