Home Intelligence Face to Face Cloud and AI are indispensable for gaming, land-based is 20 years behind

Cloud and AI are indispensable for gaming, land-based is 20 years behind

September 15, 2023

The casino industry is facing a turning point, with data as king. Earle G. Hall notes that any land-based operator who isn’t utilizing the trifecta of cloud services, big data and AI are bound to be left in the dust, as online is already 20 years ahead. Both regulators and operators stand to benefit from the shift, as demographics change. And rules have yet to be written: something the International Gaming Standards Association has planned for, and is providing a roadmap to help both sides up their game.

We’re joined today by Earle G. Hall, the CEO of AXES.ai, and the chairman of the International Gaming Standards Association. First of all, I want to delve into psychology before we branch into gaming, even though they’re completely interlinked. I’ve seen many of your talks, and I’m very curious as to how gaming ties in with the “herd mentality”, specifically, within Asia. You’ve mentioned how mirror neurons that we all have, try and “harmonize us within the environment”. Do you think that happens on a casino floor? And what are the consequences?

It’s obligatory, to be honest, mirror neurons work in the unconscious, it’s something that we’re really not aware of. And it is our brain trying to synchronize with the environment in real-time, all the time. This means mirror neurons are frantically trying to make sure that we can adapt and integrate with the herd. If that doesn’t happen, negative entropy happens. And we just leave. So, if you’ve ever been in a situation where you’re uncomfortable, and you’re trying to adapt, and after a while, it happens; or worse, after a certain amount of time, you have to leave. That’s what mirror neurons do. And it is a part of every single moment, outside of solitude that we live.

How can operators drive consumers to what they want? I mean, what are the main hindrances for that right now?

I always get in trouble whenever I give my opinion, and I’m too old maybe to stop now, but I have no idea how most operators can manage player satisfaction.

I have no idea because of the fact that there are almost no systems out there in the land-based gaming industry that are providing real-time fine-granular data. It’s important to explain what fine granular is because apparently, it’s not a known term. Fine granular means you have multi-game information and you have multi-denomination information. And when you correlate the game experience with the denomination experience, with the time on device, the average time on device, the player speeding up or slowing down their clicking: which is an indication of their behavior, their fight and flight, their happiness, their satisfaction.

When you can mine and predict, in real time, how the player is feeling. Are they happy? Are they sad? Are they frustrated? Are they excited? When you can actually mine the data in real time and predict where they’re going in the next five to 10 spins, to make sure if the losing curve is too high that you can associate bonuses with it, that’s what managing a player experience is.

But to do that, you need cloud, you need big data and you need AI. And if I put those three things together, it comes up almost zero in the land-based gaming space, for the simple reason is: almost every system out there is on property, it’s running on an antiquated technology, it’s running on a proprietary language, and has no capacity to integrating all of these cutting edge tools that are coming out.

What’s the cost of that type of division, though? I mean, is that affordable to main operators? Is that only for the multibillion dollar operators? Or can it be for the SMEs, for example, or the smaller casino operators out there?

Well, once again, and like I said, I’m the worst person to talk about AXES.ai, because I don’t like self promotion and those things. So I’ll equate this to the fact that any modern big data cloudset has the same cost for one user and 1 million users.

If we had physical servers for Google right now, or other things like that, for some of their activities, we’d need millions of computers to do those equations.

The whole world moved on about eight years ago to virtual technology. And it’s called elasticity. In other words, if you’re on a Friday night, you may have to ramp up to 128 virtual servers, where on a Monday morning, that technology if it’s done right, it scales right back down to two or three.

Well, I’ll navigate through this one softly. And I’m smiling, because if you look at all of the fines that certain regulators are now, throwing around at operators; if I was an operator, my most precious data would be the behavior of my players as a proportion or as a correlation of how I can drive more satisfaction, more caching, more time on device.

But the rest, what’s the rest? It’s sharing anti-money laundering information. It’s sharing addiction information. So, there is a common movement in our industry, where big data can protect everybody from themselves and others. And I doubt if there’d be one operator in the world who wouldn’t want to share that data. One, to appease the pressure on the regulators, and number two, to be able to protect everybody in general. So they find another industry to go do bad things in.

But there’s another part of the data which really becomes your recipe, you know, like your grandmother’s spaghetti sauce or something like that. What if the algorithm that you worked on with your technology supplier, to be able to get your players to spend 10 minutes more sitting at the seat, $30 more a session, walk out and hit the big green button for ‘I’m happy, and I love my time in your joint’, as opposed to hitting the big red button.

So there’s two types of data out there. There’s player health and there’s player satisfaction.

Player health, whether it be they’re under attack from microlenders, they’re under attack some money launderers, or they’re under attack because they’re having a bad day, and they’re starting to play abnormally. That I don’t know, I fall off my chair if there was an operator who wanted to keep that data for themselves.

But on the other spectrum, I think the formidable war that’s coming is the leading edge operators out there, and I always like to call out Melco for this one, and the Avery’s (Avery Palos) of this world. You have these geniuses that are out there, polishing and polishing algorithms, because their passion is customer satisfaction. That type of data is the secret sauce of where we’re going. But don’t forget, to be able to do that, I need real-time cloud big data, so that I’m processing the entire dataset to see any skew, whether it’s positive or negative, in the player’s experience,

Well it seems like, in that circumstance, online has a bit more of an advantage as opposed to land-based. So would you say that consumer instant gratification is easier achieved online, or in a land-based scenario?

Once again, I’m really happy you’re bringing this point up, because the only difference between online and offline, I don’t even call them land-based when you ask that question, because online, and let’s go all the way back to Boss Media (later rebranded as GTech), you’d have to look that one up on the internet, but if you go back all the way to Boss Media, the Sue Schneider’s of this world, when online gaming came online, it had nothing to do with gaming. It had everything to do with real-time information, real-time payments, real-time business intelligence, to make sure that customer satisfaction is optimal.

The only reason we’re not using that term, or those terms, in the land-based or the offline industry, is because the majority of systems are antiquated, they’re proprietary, they’re closed, they can’t integrate open source, they can’t integrate third-party. They’re all split into silos.

So, the problem is not the regulators. The problem is definitely not the operators. The problem is, and it has to be called out, is that the majority of systems out there are lost leaders within the organizations that create them.

So, if you look, there are very few organizations in our industry that their primary focus is systems, or even their only focus is systems. Like when you look at SAP, what is SAP? It’s a systems company. They have several modules: start with finance, go out to production, manage supply, they have CRM (Customer Relationship Management), they have manufacturing, but they’re a systems company.

When you look in the gaming industry, and you list maybe the top 10 companies, they’re all slot machine companies.

And they’re judged by their shareholders. They’re judged by the analysts, they’re judged by the general public, on their ship share. How many machines do they push out a quarter? They’re not judged primarily by: ‘Is their system cloud? Is their system blockchain? Is their system real-time? Have they integrated artificial intelligence? Are they providing a real-time experience with respect to all of the data?’

The land based industry is literally 20 years behind online.

Do you see the current evolution of AI being any type of threat? I mean, overall, and for the gaming industry? Should there be some types of constraints which are put onto it? We’re looking at generative AI right now. And those constraints, especially in the various roles that you have, should they be done by the companies themselves? Or should they be shaped by regulators?

So I’m really happy to say that the Board of Directors of the International Gaming Standards Association has taken a very black and white stance with respect to artificial intelligence, because most of the board members are very excited, very enthusiastic and very proactive, to embrace artificial intelligence.

But at the same time, outside of gaming if we look at any consumer-evolving AI project right now, there’s the potential for manipulation, there’s the potential for fraud, theft.

There are dangers to AI. And I know this one scares people when I say it but the only thing I can equate artificial intelligence to is the nuclear bomb.

So, nuclear energy has tremendous potential for clean, cheap, infinite energy. If you look at submarines that have been going around under the ocean for years and years and years running on nuclear reactors. It’s the thing with artificial intelligence, there’s so much good, whether it be predicting where to do an incision for a heart surgery, because you have millions of correlations of data that say you need to go one millimeter to the left.

But at the same time, there is a potential for bad.

And that’s why I’m so proud of the IGSA, I’m so proud of Nimish (Purohit), our Vice-Chairman from Aristocrat. We stood up, we created an ethical AI committee, we’ve had an overwhelming response. And just today, we had our monthly meeting, and we have our first guidance coming out for the regulators and for the operators. Very simple stuff, based on education, based on awareness, and based on what you need to know to understand where the fork in the road is.

So, it’s hard for me not to be overly excited with AI because we started off 15 years ago, at AXES.ai with three very, very black and white premises that was in our first PowerPoint.

One: the world was going to be capitalist.
Two: the world was going to be cloud.
And three: the world was going to be AI-driven.

And 15 years later, it feels like we’ve been dragging the ball and the chain for 15 years, teaching, educating, advocating, screaming, crying, biting our nails.

But looking at regulators themselves, it seems like they’re always playing catch-up. Do you think that they can actually keep up with what’s happening with these trends? How can the IGSA contribute to that, aside from obviously what you’re doing within the AI – suggestions and kind of playbook; that you are placing out there?

Well, let’s be clear. There’s only one job I would never do in this industry. And that’s be a regulator.

“It is the hardest job, the most unimaginable job there is to do in our industry”
Earle G. Hall

Because you’re in an impossible position. You have a political agenda that’s not only evolving all the time, but it’s volatile all the time. And it’s chasing the news cycle, in many cases, unfortunately.

Our operators are barely coping with ways to figure out ‘How do we address the new demographics as the population ages, as the people change, as the generation changes?’.

So when you look at that triangle that I tried to build: in the middle, that’s the reason why I nudged my board of directors at IGSA.

I said: “Hey, somebody has to stand up and be the camp counselor. Somebody has to stand up and say, “Hey regulator, we got your back”. “Hey technology companies, we’re going to hold you to something”. And “Hey operators, once we figure this out, we’ll start educating you”, like the session I’m giving in September, which is like an AI 101 course. IGSA had to stand up, not because it was our core mission, but because there was nobody else to do it.

Seeing how everything has played out, did you expect it to all happen a little bit faster? I mean, did your clients also expect it to happen a little bit faster? Or are things progressing now at the same pace that you kind of thought they were going to once you started delving into the new technologies?

I’ll divide that answer the two, first of all talk about my clients.

And I say my clients, because I take it very personally, I have the honor and the privilege to work with talk with and deal with early adopters.

So, I’m talking years ago, clients asking: ‘How do we correlate this information?’ They’re used to having their information in real-time all the time. So AXES.ai has been very, very fortunate in more than 40 companies to always find, or be found, by the early adopters that already understand that cashless, cloud and AI: that is the future of the planet.

That being said, I thought for the first five years that we were just going to be on the bleeding edge, not the cutting edge. I really didn’t expect for it to last 10 years.

We have some things to blame, we had a major economic crisis in 2008. So there were some bumps in the road along the way, macro economically. But what once again, the land-based industry has antiquated systems that are closed. So, it created a gap and, once again, you can blame our operators. Our operators are held to the fire to hold a certain performance every day based on historics. Every time they take a technology or a product risk, these operators are putting their jobs at risk.

But the flip side of that coin ,because like I said once again I had an amazing meeting with a potential client today. This person wanted to dive into real-time analytics, predictability, and it was just was a one hour meeting that ended up being almost four hours. And my heart was beating, I probably won’t even sleep tonight. I was so excited to be able to talk with an early adopter. Because that person understood that real-time data equals real-time increase in profitability.

But that being said, I understand mainstream is scared of change, because of the fact that the implications of making a change in ignorance are very, very scary. And that’s why I keep telling everybody who/what AXES.ai is: Yes, we’re the pioneer, yes we were first in a whole bunch of things. But that comes with a tremendous burden of education.

I’m really worried about our industry right now because it’s like a slingshot. When everything is going to flip, you’re going to have some operators that get it, that have implemented it, that master it, that are passionate about it. And the rest are just gonna get run over.

And I don’t say that as a doomsday scenario. It’s happened in every other industry. If you look and go: ‘What happened to Kodak? What happened to Xerox? Hey where’s Compaq today?’

And I could go on forever with my list. But the thing is, is that the more tension, the more static electricity you put on a problem, the more you hold back the change, the more brutal that correction is going to be.

And like I said, we work with the early adopters, it’s absolutely the funniest thing I’ve ever done in my life.

But for the mainstream, land-based, industry, it’s going to be quite a shock when an operator in Las Vegas adopts cloud, adopts AI, adopts predictability, and their numbers start to go up. And it’s only two, three, four, five, even six quarters later that their competitor wakes up and goes: ‘Why are my numbers constantly trending down?’

But then the cycle to get out of server-based technology, to get into serverless cloud, to get into real-time data collection, and to get to artificial intelligence, they’ve lost three years of their lifecycle.

I’m not sure in the years to come, organizations are going to have three years to correct.