Artificial intelligence is by no means a new invention, and it already has a significant impact on the way we live our lives. The Internet is already largely driven by artificial intelligence, and we’re seeing algorithms and applications become more sophisticated by the day. Even our home lives are now benefiting from artificial technology, with the advent of digital assistants and smart technologies, and in the near future, this is a trend that will only continue to develop and grow.
There’s lots AI can do well, and most people are agreed that it has tremendous potential to benefit humankind. But when it comes to probability, say in the context of gambling, can AI help you beat the odds and bring the house down? Or is this a step too far for a technology that despite its age and progress, still remains dependent on the availability of information to produce effective results?
Notionally, of course it’s possible for AI and robotics to beat probability. In a purely academic sense, AI allows for infinite applications, and we see in the likes of brute force attacks and hacks this method working to good effect. Take the example of a Google search at its most basic level – Google is able to compute even complex information in infinitesimal time periods, performing tasks that might otherwise have taken days or even weeks to research.
The same applies across computing generally, where computers are programmed to handle incredibly complex strings of data, and to process that information and sometimes even make decisions based on the outcomes, all in fractions of a second. So it stands to reason, that given the right input, AI can indeed help beat probability, and give the individual controlling the AI an edge.
Back in the 60s, a scientist by the name of Arthur Samuel was already experimenting with AI in the gambling sector, putting these theories to the test. He developed an application that successfully beat the top ranked American draughts player, in a stunning demonstration of what was possible with AI and programming – even back then, almost 60 years ago.
More recent technologies in this area have achieved similar results, like the poker-playing supercomputer Claudico, which took until 2015 for developers to get right. In these respects, it’s clear that the technology has the capacity to compete with, and even outperform, expert humans in games of probability, chance and even skill.
In the physical casino environment, few would argue that this technology can and does help shorten the odds of a successful outcome, and gives players a competitive edge they wouldn’t otherwise have had.
However, it is important to draw a distinction here between offline and online casinos. Due to the unique construction of the latter, there are some problems for AI in its ability to beat probability and decrease the house edge of given online casino games.
Take the example of online slot machines such as Book of Dead. In order for AI to beat such a game, it would have to have access to all available information. In draughts, for example, the information is contained, obvious and visible, and it is readily possible for AI to interpret and make decisions on the strength of that input.
RTP is calculated across several unknown variables, such as a number of spins, a number of players, different staking types, a number of different brand operators, and so on. These figures are all factored in and computed to calculate an average, but only the answer is available. As a result, AI cannot extrapolate enough meaningful data to improve the odds and beat probability, at least as far as online casinos are concerned.
AI is clearly a significant step forward for humanity, and it is already a hugely impressive field of technology, getting better by the day. While it may not be the most effective tool for beating the odds or probability, AI can still be used by unscrupulous elements to break games, for example, or to otherwise provide an unfair criminal advantage to those that seek to exploit it.
As a result, casino operators, as well as most other online businesses, need to be constantly mindful of the threats posed by AI – even if these don’t include an ability to skew probabilities against the casino in the online environment.