Massachusetts Gaming Commission Presses Operators Over AI Use In Marketing
The MGC has issued a detailed questionnaire to every licensed sportsbook asking whether algorithms single out losing bettors for promotions, with responses due November 3.
The Massachusetts Gaming Commission has formally begun its inquiry into how sports betting operators deploy artificial intelligence in customer marketing, sending every licensed operator a lengthy set of questions it wants answered within weeks, according to Legal Sports Report.
Operators have been asked to supply more than 70 individual responses by November 3. Taking part is voluntary, and the MGC's letter states that answers will be treated confidentially and fed into future AI policy work rather than published or used for enforcement.
The exercise traces back to reporting by The New York Times, which alleged that DraftKings had used artificial intelligence and machine learning tools to identify customers who were losing money and direct promotional offers at them to keep them wagering. DraftKings disputed that characterisation when approached by The Times.
Rather than confine its scrutiny to DraftKings, the MGC said last month it would examine AI marketing practices across the entire licensed market in Massachusetts.
The state already prohibits operators from deploying AI systems that are, in the regulator's wording, known or reasonably expected to make a gambling platform more addictive. MGC Commissioner Paul Brodeur said last month that Massachusetts appears to be one of very few states, possibly the only one, to write AI specifically into its gambling regulations.
What the letter actually asks
Before listing its questions, the MGC set out a broad definition of artificial intelligence, one that captures any automated system used to rank, classify or generate predictions about customers, regardless of whether an operator brands the tool as personalisation, optimisation or risk assessment.
The questionnaire runs across eight sections. The opening two cover general AI deployment and the underlying data these systems draw on. The remainder probe how promotions are decided, who receives them, when they land, and whether customers who have been losing money are handled any differently from others.
Among the questions operators must answer: whether a patron's losses feed into marketing decisions, whether sudden shifts in betting behaviour such as bigger wagers or more frequent deposits trigger promotional targeting, and whether a customer who has recently lost heavily is treated differently from one with a similar wagering pattern but smaller losses. Operators that say no to being asked whether rising losses or other signs of harmful play make a patron more likely to receive an offer must also describe the safeguards that prevent this from happening.
The final section turns to human oversight, asking at what stage a person reviews or can override an AI-generated recommendation, and whether staff are given an explanation for why a model flagged a given customer for a given offer. It also asks operators to show they could identify the data and logic behind any single promotional decision if a regulator queried it, and whether they can calculate what share of a customer's promotions were shaped by AI or machine learning.
MGC Chair Jordan Maynard said last month that "any further action or policy making can then be determined by the commission when and if appropriate."
Legal Sports Report reported that the Michigan Gaming Control Board said it is reviewing the issues raised by The Times' reporting, though it has not committed to a review focused specifically on DraftKings. Michigan's gambling statutes and regulations currently contain no provisions addressing artificial intelligence.