technology · artificial intelligence
DraftKings' AI Scored Bettors by Predicted Losses to Target Promotions, NYT Investigation Finds
Published · Updated
Cited: The New York Times, Gadget Review, Tech Times
TL;DR
A New York Times investigation found DraftKings built an AI model that scores bettors by predicted losses to target promotions, while shelving a parallel tool that could have flagged problem gamblers.
Why now
The investigation arrives as problem gambling rates hit record highs and Congress considers the SAFE Bet Act to ban AI personalization in sports betting.
Agree / conflict
DraftKings disputes the characterization, saying promotions go to engaged users, not those ranked by losses—but former employees and data analyst testimony suggest the model inherently targets problem gamblers.
Takeaway
Free bet offers may be algorithmically calibrated to exploit addiction markers; users should proactively set deposit limits and self-exclusion to counteract targeting.
DraftKings’ promotional AI, trained on thousands of casino players, selected features—playing frequency, daily balance patterns, loss-to-wager ratios—that are clinically linked to disordered gambling. By optimizing for post-promotion losses, the model effectively learned to rank users by addiction markers. The same data pipeline could have been redirected to detect at-risk behavior, but DraftKings shelved that project, citing insufficient evidence.
FAQ
How did DraftKings’ elasticity score work?
The model analyzed each user's playing frequency, daily account balance patterns, and loss-to-wager ratio to generate a score predicting how much they would lose after receiving a promotional offer. Higher scores meant the user was expected to generate more revenue per promotional dollar spent.
Did DraftKings also build a tool to detect problem gamblers?
According to the NYT investigation, a data scientist began developing a model to identify customers sliding toward a gambling crisis, using the same infrastructure. DraftKings shelved the project, with its chief responsible gaming officer saying the approach was not sufficiently "evidence-based."
What is the SAFE Bet Act?
Introduced by Sen. Richard Blumenthal and Rep. Paul Tonko, the bill would prohibit sportsbooks from using AI to track individual betting habits, generate personalized promotional offers, or develop AI-driven betting products. It remains pending in Congress.
How can bettors protect themselves?
Users can set deposit caps, time limits, and self-exclusion through platforms' responsible gaming centers. Reviewing account loss history via tools like DraftKings' My Stat Sheet can provide transparency. The National Problem Gambling Helpline (1-800-MY-RESET) is available 24/7.
Sources
Canonical URL: /trend/2026/draftkings-ai-scored-bettors-by-predicted-losses-to-target-promotions-nyt-invest
Disclaimer
Digests summarize public sources. They are not advice, forecasts, or a complete record of every trend.
This digest was compiled by Yanuki using publicly available data and trending information. The content may summarize or reference third-party sources that have not been independently verified. While we aim to provide timely and accurate insights, the information presented may be incomplete or outdated.
All content is provided for general informational purposes only and does not constitute financial, legal, or professional advice. Yanuki makes no representations or warranties regarding the reliability or completeness of the information.
This digest may include links to external sources for further context. These links are provided for convenience only and do not imply endorsement.
Always do your own research (DYOR) before making any decisions based on the information presented.