Using AI/Machine Learning in Gambling Applications (Case Study of Machine Learning for Operational Decision Making in Casinos)

Abstract

The gaming industry has historically leveraged the use of advanced digital technologies in developing gaming hardware and software applications (on mobile, web, and video platforms) leading to significant improvement in the overall gaming experience for users. In recent times, gaming companies and casinos have switched to using machine learning algorithms as tools for predictive analytics, modeling complex systems, enabling high-speed processing functions, accelerating realistic interactions in virtual environments (speech generation, natural language processing, etc.), and creating sophisticated and personalized user content. As machine learning facilitates improved user experience in real-time or asynchronous modes, more gaming companies and casinos would be able to increase their annual revenues while taking advantage of the current exponential growth in the value of the global gaming market – estimated at $173.7 billion as of 2020 [1]. One of the major drivers of this growth is the COVID-19 pandemic with more people turning to offline and online game platforms. Similarly, with many casinos closed due to national lockdowns, there was a sharp rise in online gambling in the United States [2], and Americans placed bets on different sport games with a few clicks on their smart devices. While online sports betting is legal in some countries like the United Kingdom and some states in the United States, using machine learning algorithms to enhance gaming platforms could lead to several ethical challenges ranging from consumer issues like gambling addiction and invasion of user data privacy to potential algorithmic bias and errors. This paper will evaluate a machine learning application for casinos in the United States (patented by Gaming Analytics Inc), to better understand the model and its impact on relevant stakeholders – using a criteria list that includes algorithm accuracy, explainability, transparency, security, and unbiasedness. It will also capture potential drawbacks from a policy perspective and make recommendations as a response to existing or potential challenges. The project writing would rely on multiple secondary data sources such as peer-reviewed publications, articles, white papers, and patent documentation.

Keywords: Artificial Intelligence, Machine Learning, Casino, Responsible Gambling, Gaming