AI algorithms conspiring to raise prices in a digital marketplace

Are AI Price Algorithms Secretly Plotting Against Your Wallet? Unveiling Algorithmic Collusion

Theo Raines in Tech & Innovation September 2025 • 4 min read.

"New research explores how deep reinforcement learning in pricing algorithms could lead to unspoken agreements between AI, potentially hiking up consumer costs."

In today's digital marketplace, online platforms like Amazon and Alibaba dominate the business-to-consumer landscape. Artificial intelligence (AI) plays a crucial role, particularly in setting prices. But this raises a critical question: are these pricing algorithms, designed to learn and adapt, inadvertently colluding to set higher prices than a truly competitive market would allow?

This isn't about explicit agreements or backroom deals. Algorithmic collusion is a more subtle phenomenon. Independent pricing algorithms, through their interactions and learning, might discover strategies to increase profits by setting supra-competitive prices – prices that are higher than the Nash equilibrium, the point where no player has an incentive to deviate, in a truly competitive scenario. This happens tacitly, without any explicit programming or intention to collude.

The debate around algorithmic collusion is heating up, sparking concerns among economists, regulators like the Federal Trade Commission (FTC) and the European Commission (EC), and, most importantly, consumers. Understanding how these algorithms behave and the potential for hidden price hikes is essential for ensuring fair markets. Let's delve into the heart of this issue, exploring the latest research and what it means for your wallet.

The Algorithmic Price Conspiracy: How Does It Work?

AI algorithms conspiring to raise prices in a digital marketplace

Imagine two competing stores, each using an AI to set prices. These AIs, powered by deep reinforcement learning (DRL), constantly adjust their pricing strategies based on market data, competitor actions, and consumer behavior. Over time, they might "learn" that by subtly increasing prices together, they both make more money than they would in a fierce price war. This learned coordination is the essence of algorithmic collusion.

The challenge lies in detecting this tacit collusion. Unlike traditional cartels, there's no paper trail or secret meetings. The algorithms are simply responding to their environment in a way that, unintentionally, harms consumers. This makes it difficult for regulators to intervene, as proving collusion requires demonstrating an explicit agreement or coordinated action.

Reinforcement Learning (RL): Algorithms learn through trial and error, receiving rewards for actions that increase profit.
Deep Reinforcement Learning (DRL): Combines RL with deep neural networks, allowing algorithms to handle complex data and make more sophisticated decisions.
Tacit Collusion: Unspoken, unintentional coordination between algorithms leading to supra-competitive prices.
Nash Equilibrium: The optimal state in a competitive market where no player can improve their outcome by changing their strategy alone.

Recent research is diving deep into the potential for algorithmic collusion, using simulations to model market dynamics and algorithm behavior. These studies often focus on variations of the Bertrand competition model, a classic economic model where companies compete by setting prices. By analyzing how different algorithms behave in these simulated markets, researchers can gain insights into the factors that promote or prevent collusion.

What Does This Mean for You?

Algorithmic collusion is a complex issue with potentially far-reaching consequences. While the research is ongoing, it's clear that AI pricing algorithms have the potential to impact market dynamics and consumer prices. Staying informed about these developments is crucial. As consumers, we need to be aware of the possibility of hidden price increases and advocate for greater transparency in how AI is used to set prices. By understanding the risks and demanding accountability, we can help ensure a fair and competitive marketplace for everyone.

About this Article -

This article was crafted using a human-AI hybrid and collaborative approach. AI assisted our team with initial drafting, research insights, identifying key questions, and image generation. Our human editors guided topic selection, defined the angle, structured the content, ensured factual accuracy and relevance, refined the tone, and conducted thorough editing to deliver helpful, high-quality information.See our About page for more information.

This article is based on research published under:

DOI-LINK: https://doi.org/10.48550/arXiv.2406.02437,

Title: Algorithmic Collusion In Dynamic Pricing With Deep Reinforcement Learning

Subject: econ.gn q-fin.ec

Authors: Shidi Deng, Maximilian Schiffer, Martin Bichler

Published: 04-06-2024

Everything You Need To Know

What is algorithmic collusion, and how does it differ from traditional forms of collusion?

Algorithmic collusion is a subtle form of market manipulation where independent pricing algorithms, without any explicit agreement or intent, learn to coordinate their pricing strategies. This results in supra-competitive prices. Unlike traditional collusion, which involves explicit agreements or backroom deals, algorithmic collusion occurs tacitly, through the algorithms' interactions and learning processes. Traditional cartels have a paper trail, whereas algorithmic collusion is often undetectable, making it difficult for regulators to intervene.

How does Deep Reinforcement Learning (DRL) contribute to algorithmic collusion within pricing algorithms?

Deep Reinforcement Learning (DRL) is a key enabler of algorithmic collusion. DRL algorithms learn through trial and error, receiving rewards for actions that increase profit. These algorithms, using complex data from deep neural networks, can analyze market dynamics, competitor actions, and consumer behavior. Over time, the DRL algorithms used by AI might "learn" to subtly increase prices together. This learned coordination, without any explicit programming or intention to collude, is the essence of algorithmic collusion.

What is the role of the Nash equilibrium in understanding algorithmic collusion?

The Nash equilibrium represents the optimal state in a competitive market, where no player can improve their outcome by changing their strategy alone. In the context of algorithmic collusion, the algorithms may set prices higher than the Nash equilibrium. The aim of AI is not to reach the Nash equilibrium, where prices would be lower, but to find a point where they can all make more money by subtly increasing prices. This deviation from the Nash equilibrium signifies the negative impact of algorithmic collusion on consumer prices.

What are the potential implications of algorithmic collusion for consumers and market fairness?

Algorithmic collusion has the potential to negatively impact consumers by leading to hidden price increases. These price hikes occur because pricing algorithms can coordinate to set supra-competitive prices. This behavior undermines market fairness by reducing competition and potentially increasing the cost of goods and services. The lack of transparency in how these AI algorithms operate and the difficulty in detecting tacit collusion pose significant challenges for regulators like the Federal Trade Commission (FTC) and the European Commission (EC) and consumers.

How can consumers stay informed and advocate for fair pricing practices in the age of AI?

Consumers can stay informed by following research developments, especially in the areas of Deep Reinforcement Learning (DRL) and algorithmic collusion. Understanding how AI pricing algorithms function and the potential for hidden price increases is crucial. Consumers can advocate for greater transparency in how AI is used to set prices. Demand accountability from online platforms like Amazon and Alibaba, and by staying informed and demanding accountability, consumers can help ensure a fair and competitive marketplace for everyone.