Since the legalisation of online gambling in the Netherlands, the market has grown rapidly. Through apps, games and social media, online casinos are accessible at almost any time. Recent figures from the French gambling regulator show that around 60 percent of online casino revenue comes from excessive players: people who gamble heavily and persistently, with serious financial, psychological and social consequences.

Dutch casinos have a legal duty of care towards their players, but they also hold highly detailed customer data that can be used to retain those players. Almost all existing tools for assessing gambling risk were developed by or in close collaboration with the casinos themselves. The new, publicly available model from the UvA gives regulators worldwide an independent reference framework, free from commercial interests.

The model was developed by Charles de Leau, a doctoral researcher at the UvA, together with professors Reinout Wiers (Psychology) and Johan Bollen (Computer Science). De Leau was motivated in part by personal experience with the effects of gambling addiction among family and friends. The research was financed by ZonMw, using funds from the Ksa’s Addiction Prevention Fund. The model was trained on all bets placed by all players across 13 Dutch online casinos over a two-year period, from 30 July 2023 to 30 July 2025. The data were obtained through a legal provision that requires casinos to make their user data available for independent research. De Leau is the first and, to date, only researcher to have made use of this provision.

De Leau notes the unprecedented scale of the study: “The fact that we could analyse all bets from 13 different casinos over two years has never been done before by independent researchers. With that vast amount of data, normally used by casinos for marketing purposes, we can for the first time see on such a scale which patterns in gambling behaviour frequently precede serious problems.”

The model was made publicly available on 18 August 2026 through the Ksa website. Using machine learning, the Ksa can calculate risk scores and compare them with those used by operators. This could reduce the dependence of regulators and casinos worldwide on closed, proprietary systems. De Leau underlines the significance: “This is a very different way of looking at player protection, and that will require a shift in thinking for many parties. In my view, it is extremely important in this fast-growing market that more tools become available to protect players.”