Conference proceedings, Zurich 2023

detection, automate more of the of the handling of the messaging, and then allocate the human resources on RG to the most severe cases or more specific things, bearing in mind that some of our customers have one, two million active players today, then a very small percentage of new actions is still a lot to handle. That’s part of this: working smarter, not harder and automating as much as possible. EH: False positives are only coming from two places. They’re coming from bad data, or bad algorithms. There’s only two places it can come from. If you’re looking at land based, 90% of data is bad. How’s that for a number? I think we’re all going to have to lean in and rely a little bit more on a future GLI that’s going to be pointing the finger at data and pointing the finger at algorithms. If I was running a jurisdiction, I would never let an AML certificate go directly automated out into the ether. Why? Because in the next five years you will see the amount of litigation that’s going to be caused by automated processes that don’t have an authority sign it. What’s going to happen? Are you going to take the algorithm to court? IH: I think one of the things to be wary of is to know that that data can be flawed. The data that we see from devices now is being used for a purpose or intention that it was never really intended for. It was intended for tax collection and understanding revenue. AI is not just in the gaming industry, it’s across all industries. We see it on airlines on transportation on medical on that banking. And there is very little work on the technical standards. There are some five pillars that we’ve identified as being fundamental to AI. They are inclusive growth, human-centered values and fairness need to be one of those cornerstones, transparency and explainability, which the previous panel touched on, robustness, security, safety, integrity and accountability. Because when an AI makes a decision, there has to be some justification otherwise you just have software developers pumping out algorithms that may come up with unintended consequences. So this is a very new area and lots of risks and benefits that are out there for the industry. Questioner: In the gaming industry there is a lack of security and we know, there’s a danger of cheating, theft of money and identity. But what about in the real world right now? What have been the consequences of having a lack of security? Like having the server servers sitting on a casino? EH: Bravo, sir. One of the biggest risks when you’re collecting this data to monitor for Responsible Gaming or for AML you can also use the other way to actually abuse of that data to find who is more, let’s say susceptible, who was more easily identified as the right victim for black market, for example, gambling. So if I were a gambling operator in the black market and I get the data, I would use it to attract the players that are more at risk of falling Problem Gambling. IH: that gets us into a whole new area and one for another day. For now, though, let’s just thank our panelists.

62

Made with FlippingBook flipbook maker