DraftKings customers allege AI targeting as state scrutiny widens

Daniel Vest says he received seventy promotional messages in a single month. He was losing money. The messages kept coming. That fact sits at the center of a proposed class action filed in the US District Court in Boston, and it is the detail Massachusetts Attorney General Andrea Campbell's office chose to highlight when her staff said the allegations raise serious concerns about the use of technology to target consumers vulnerable to problem gambling.

The lawsuit, filed by Vest on behalf of a proposed class of DraftKings customers, rests on a New York Times investigation that drew on accounts from former employees. Those employees described machine-learning systems designed to identify losing bettors most likely to respond to promotional incentives. A separate thread in the complaint: DraftKings allegedly explored technology to flag customers showing signs of gambling disorders, then stepped back from those efforts. Whether stepping back matters legally depends on whether the initial exploration constitutes a violation of Massachusetts consumer protection law — a question the court has not yet reached.

DraftKings has denied the allegations directly. Their public statement is unambiguous: no AI targeting based on losses, no AI targeting based on problem gambling indicators. The denial is clean enough that it narrows the litigation to a factual dispute about what their systems actually did, which is where these cases almost always end up grinding for years.

State Auditor Diana DiZoglio added her voice to Campbell's, saying the accusations would be unacceptable if proven true. Two statewide officials commenting before any finding of fact is a signal worth reading — not about the merits of Vest's case, but about the political environment DraftKings is now operating inside in Massachusetts.

I have watched companies face this pattern before. A journalism investigation, a named plaintiff, elected officials stepping in front of cameras. What happens next almost never tracks the legal merits in the short term. It tracks the regulator's appetite and the company's willingness to move faster than the litigation forces them to. DraftKings is in the denial posture, which is defensible if their systems are as described, and costly if discovery produces documents that complicate the story.

The piece of this I don't think is getting enough attention is the Massachusetts law angle. Vest's complaint alleges the company violated state law by failing to disclose its AI use to customers. That is a different theory than "you targeted problem gamblers." It is a transparency claim, and transparency claims are harder to defeat on the facts because the question is not what the AI did but whether customers were told anything at all. Disclosure cases tend to settle. The question is at what number.

I am adjusting here for my own tendency to find the worst case. The disclosure claim is real, but the class certification fight in AI-targeting cases has been difficult for plaintiffs, and Vest's individual harm — money lost while receiving promotions — does not automatically translate to a class with unified claims. This goes further than the June Chicago lawsuit did, because the New York Times investigation gives it a factual foundation that the earlier case lacked. Whether that foundation survives DraftKings' discovery responses is the open question the market on this outcome should be sitting with.