A rising tide of fraud is being evidenced across numerous facets of life, with the gambling industry certainly susceptible to being struck by such criminality.
Those operating on the other side of the law are rapidly evolving their methods when it comes to fraud innovation, as well as the techniques being deployed regarding issues such as AI fabrication.
In conversation with SBC News, Warren Russell, Founder and CEO of eyeDP, explores this upsurge, looks at what AI can do to help prevent this new generation of fraud, and details how operators and businesses need to be one step ahead of criminals.
SBC News: In your opinion, does the gambling industry make full effective use of AI technology when it comes to combatting a range of ongoing issues?
Warren Russell: I think what the gambling industry has been good at is using AI for customer personalisation, marketing, game optimisation, and these kinds of things.
I think that they’ve been quite poor in implementing it across back office systems, but I don’t think that’s an issue with just gaming per se. I think that’s across most sectors. They go for the snazzy revenue generating stuff before plugging the hole in the bottom of the bucket.
There are of course companies that use advanced technology around duty of care, Crucial Compliance and MindwayAI, for example, but even specialists such as those are only just starting to scratch the surface of what is possible. On the whole, I think that it’s far less widely adopted in those core protection parts than it is in the wider gambit.
SBCN: Is it fair to say that in terms of adapting and evolving, criminals are more savvy than the industry itself?
WR: Absolutely. The ethos that I’ve always lived by, having worked in RegTech for 15 years now, is that you’re always playing catch-up with criminals. They’ve got no regulations or laws to abide by, they don’t have to have an ethics policy or anything like that.
The problem now is that AI has turbocharged fraud. I did a bit of research and UK Finance state that there were over two million confirmed cases of fraud in the UK in the first half of this year, which is almost a 20% year on year increase. A lot of that is driven by the increase in deep fakes and synthetic IDs making it harder to detect.
There’s a lot of solutions out there to solve parts of the problem, I mean, part of what we do is to look for evidence of tampering or inconsistencies in documents. But if you are creating a document or an identity from scratch, you can make it more consistent and plausible, and it starts to become more challenging to detect. There’s a lot to gain for criminality in adopting technology quickly.
SBCN: Fraud continues to be a prevalent issue across numerous jurisdictions. Just how serious do you believe this issue currently poses to the industry?
WR: It’s getting worse and bigger because it’s getting easier – bonus abuse, identity fraud, money laundering, multi accounting. There’s so many different avenues now that can be hit again.
Gambling fraud grew by 64% between 2022 and 2024. The baddies are getting more active and getting more success, but the repercussions are also getting bigger for the operators if they’re not implementing controls.
We’ve had more fines announced recently for household name operators. It’s an issue both from the loss of money, loss of license, loss of reputation, everything. It’s like we’re in a bit of a perfect storm at the moment.
SBCN: How can AI be better utilised to help prevent this escalating even further?
WR: When I set up W2 in 2011, orchestration wasn’t a thing. I think that has now become more the norm and it has certainly helped. Businesses are starting to unify their data and systems rather than having siloed processes. This is where I think the developments in AI can really take it to the next level.
A lot of the orchestration platforms do a really good job of making it easier for operators to consume services. ‘I’ve only got to integrate to one provider and I’ve got access to 40’, rather than having to all integrate to 40 providers.
I can now see my customer journeys more clearly which means I can implement more detailed analysis. This is where technology can really make a difference – looking for the inconsistencies, looking for the links, joining the dots in between all those elements that are now pulled together.
SBCN: What have you identified as the key techniques being adopted by criminals for AI fabrication, and how can these be best prevented?
WR: When you talk about the deepfake videos, synthetic IDs and fabricated documents and all of these elements, what you need to understand is that you have to use AI to catch AI.
It’s a little like what’s been used in academia for a long time with plagiarism detection. You’ve got to use the data and tools that the people who are ripping off others are using in order to catch them out. There’s some good platforms out there that are really good at picking up on this stuff.
AI’s been around for years, depending on the way you look at it. I’m very reluctant to say AI can solve it, but AI is another important tool in the armory.
I think the whole social engineering piece is where it really becomes powerful, with the ability to scour and research millions of sources of data in seconds.
We’ve all stuck stuff into ChatGPT, or something similar, when we’re trying to work something out, and it’s ability to go out and look across a massive real estate. That’s where I think we’ve really got to use the data, because there is always a chink in the armor, but you’ve got to search everywhere to try and find it.