Key takeaways:
- The Department of Justice (DOJ) has again emphasized that software pricing tools relying on nonpublic data in algorithms and AI could present meaningful criminal antitrust risk.
- A company’s use of commercially available software does not reduce its exposure and may increase scrutiny, particularly where tools incorporate competitors’ sensitive data or generate coordinated outcomes.
- Companies should ensure that software pricing tools are subject to robust compliance controls, data governance and employee training.
At a recent conference, DOJ Antitrust Division Criminal Deputy Daniel Glad reiterated that price collusion via software tools remains a continued priority for criminal enforcement. Although pricing algorithms may appear novel, Glad emphasized that the “mechanics of collusion” are not and remain subject to the same legal standards and evidentiary burdens as traditional antitrust cartel conduct. The Division believes that the Procurement Collusion Strike Force and the recently established Whistleblower Rewards Program have primed the DOJ to address these developing issues. Even though the Whistleblower Rewards program was not designed with AI in mind, Glad stated that “it was designed for exactly this kind of detection problem, and it is well positioned for the risks that AI and algorithmic tools will create.”
The central compliance takeaway is that software tools, even if commercially available and well established in an industry, do not insulate companies from antitrust risk. “Software does not change the rule. And it does not soften the consequences when the rule is broken,” according to Glad. To the contrary, their use can enhance enforcement visibility by generating extensive structured records, including data inputs, queries, logs and time stamps for parameter changes. Each of those digital artifacts may be used to detect and prove potential collusive behavior.
Glad explained that criminal liability may arise where competitors “understood that their sensitive nonpublic data will be used to set prices for competitors and have participated on that understanding.” In that case, “the door to the per se rule, and therefore to criminal enforcement, is open.”
Glad identified several factors that may indicate increased antitrust risk, including where competitors adopt a common algorithm, where tools rely on competitors’ nonpublic data, where users understand their data will influence competitors’ pricing and where outputs are expected to be followed across the market. At the same time, the DOJ has recognized that certain uses of aggregated, anonymized and sufficiently aged data may be permissible, underscoring the importance of data governance in compliance assessments.
The DOJ is also focused on distinguishing lawful “vertical” software arrangements from unlawful “hub-and-spoke” conspiracies. There, the key question is whether a provable agreement among competitors – a “rim” – sufficient to support a horizontal conspiracy may exist. But an agreement may not necessarily involve direct communications with competitors; rather, particular emphasis will be placed on how tools are used and understood by market participants, not just how they are structured.
In addition to proving an agreement, criminal enforcement requires proof of intent. Glad emphasized that in this context, “intent travels with the human decision to contribute to and rely upon the system.” As a result, compliance programs should address not only technical inputs and outputs but also how and why tools are marketed, implemented and relied upon in practice.
Glad further cautioned in-house counsel to institute compliance programs. “[I]f your company is deploying AI or algorithmic tools in competitively sensitive areas, then antitrust review cannot be ceremonial. … It must be real.”
This warning is consistent with the Antitrust Division’s November 2024 update to its Evaluation of Corporate Compliance Programs in Criminal Antitrust Investigations to include risk assessment questions specifically directed at AI and algorithmic tools. As Glad stated, “[I]n short, [the compliance updates] ask whether a company’s risk assessment addresses its use of new technologies, particularly AI and algorithmic revenue management software; whether a company assesses antitrust risk as new technology tools are deployed; whether compliance personnel are involved in that deployment; and what steps the company is taking to mitigate risk.”
Practical Compliance Considerations
Companies developing or using pricing software tools should consider:
- Maintaining visibility into algorithmic tools used in pricing, procurement, bidding and compensation decisions
- Assessing what inputs may be used by the algorithmic tool, including whether competitors’ sensitive data is incorporated into training and/or outputs
- Evaluating outputs from the algorithmic tool and any business reliance on the same, particularly where certain recommendations may consider competitors’ pooled data
- Training relevant personnel to recognize that antitrust risk can arise from routine use of commercially available tools
[View source.]