AI in the Gaming Industry

AI is seemingly everywhere these days and so it stands to reason that it has made its way into the gaming industry as well. Given the amount of data that online gaming generates, it again makes sense that online suppliers and operators would be the first to adopt this technology. Analyzing millions of records and making sense of them is done more efficiently using a Large Language Model (LLM) algorithm. AI algorithms are great at identifying anomalies and trends. Using this data, operators can better tailor marketing campaigns, suppliers can hone it on the qualities that drive player engagement, and everyone’s happy. Right?

Well, just with most things in life there is a catch. With AI the catch is the data.

AI algorithms are data hungry beasts that require large data sets to deliver results. The good news is that the online gaming sector has lots of data. The bad news is, however, that it may not be clean, accurate, complete and unbiased data. The old adage of Garbage In – Garbage Out, has never been more true.

While regulators are trying to understand the AI algorithms- a bit of a fools errand when it comes to models using generative AI where the algorithm changes automatically as it learns -they should be focusing on the data. That’s not to say that they should not have an understanding of the algorithm. It’s just that they don’t need to go down to the zeros and ones. We understand and applaud regulator’s’ efforts in this area. They want to make sure that these AI algorithms and the applications in which they are embedded, are fair and unbiased. They are also concerned about data privacy. These are all very worthwhile concerns! The issue is that trying to deal with AI at the technological level is a very difficult and potentially unachievable objective. We believe regulators, and by extension the industry, would be better served if they focused on the data. What does that mean?

Let’s start with the big questions that must be answered:

• What data should AI Algorithms be allowed to consume?
• Whose responsibility is it to ensure the data is clean, accurate and unbiased?
• Whose responsibility is it to document the intended purpose and outcome of AI algorithms?
• Whose responsibility is it to monitor the AI algorithm and stop its use if the outcome deviates from its intended purpose?
• Whose responsibility is to notify regulators, affected individuals, etc. to ensure transparency and garner trust?
• What is the time frame by which notification is to happen?
  

These are some of the issues being tackled by the International Gaming Standards Association’s (IGSA) Ethical Use of AI committee. Their Best Practices take a common sense, pragmatic approach towards the use of AI, providing regulators a base set of requirements that they can use to help them provide oversight without stifling innovation, ensure fairness without overburdening regulation, and focus on what is truly important when it comes to AI and Big Data – DATA!

For more on IGSA’s Best Practices check out their website at http://www.igsa.org