Massachusetts regulators will examine how DraftKings and the state’s other licensed sports-betting operators use artificial intelligence and machine learning, beginning with the company after allegations about its promotional targeting.
The review will cover customer acquisition, promotions and responsible gaming. Chairman Jordan Maynard said the commission was concerned by the application of rapidly evolving AI technologies and had asked Executive Director Dean Serpa and staff to engage with DraftKings to establish the specifics reported.
The commission had not determined that Boston-based DraftKings committed wrongdoing when it announced the inquiry. The New York Times investigation alleged that the company directed employees to build a machine-learning model that identified customers most likely to respond to promotions and then lose money. It drew on interviews with more than 40 former employees, internal communications and customer betting records.
DraftKings told the Times that it rejected any implication that its marketing was unfair or improperly targeted customers. Massachusetts rules already bar operators from using AI to target promotions known, or reasonably expected, to make a platform more addictive.
The initial engagement with DraftKings will be followed by an examination of AI use across all licensed operators. Serpa and the commission’s AI task force will conduct the work, and the commission could consider additional guidance or regulation after reviewing the operators’ practices. Commissioner Paul Brodeur called the Times report “troubling” and said regulators needed to establish the facts on the ground.
The task force follows a one-year research project awarded by the commission to UNLV’s International Gaming Institute in July 2024. Its November 2025 report examined AI’s current and potential gaming applications, with an emphasis on player-risk and financial-risk identification.
That study described advanced personalization as a “double-edged sword”: it can improve engagement while creating ethical risks in targeting vulnerable people. It found gaps in rules governing gambling-specific AI uses, including marketing, personalization and behavioural nudging, while also noting that proprietary risk-detection systems often lack methodological transparency.
The UNLV researchers recommended that gambling regulators appoint an internal AI champion or task force to monitor licensed operators’ technology use. Operators, meanwhile, have argued that AI can help identify potential problem gamblers.
