March of the Machines
Published 19th December 2015
Artificial intelligence (AI) is a broad subject area, one that if were you ask the public about, the answer that would likely be front of mind is that of a physical robot that has consciousness. The reality however is that it isn’t the robot, but the computer program, that possesses the intelligence. For example, the software behind Apple’s virtual companion application Siri is AI, and the woman’s voice you hear is a projected personification of that AI, with no physical robot component involved. AI and its applications for us mere mortals can be summarized in three usage cases or categories.
The first is perhaps best referred to as ANI or Artificial Narrow Intelligence, also sometimes referred to as weak AI. This means the AI specializes in one particular area of application and doesn’t have genuine intelligence capabilities. A good example of ANI is our previous example of Siri or when you contact your credit card provider or bank, and you talk to the automated voicerecognition service that can never understand your accent. The second category is Artificial General Intelligence (AGI). This is often referred to as strong AI or humanlevel AI, in other words, the machine can perform human-like intellectual orientated tasks, such as problem solving and to think abstractly. An example of AGI often cited is IBM’s Deep Blue chess-playing computer, which gained notoriety by beating then-reigning world chess champion Garry Kasparov in game one of a six-game match. The third category is Artificial Superintelligence (ASI). This is the computer that is much smarter than the best human minds in every academic field; it’s the super-computer that can evolve into a higher state of consciousness’ and of transcendence.
AI and iGaming
We know we are sometime away from selfaware AI applications such as the fictional Skynet, or so big brother tells us. But what are the day-to-day applications for the iGaming industry? Well, most of the applications involve machine-based learning programs. This is where, yes a human, writes a series of algorithms that looks for patterns in data. The objective of machine-based learning programs is to use data to improve the program’s understanding and adjust the program outputs in real-time. A good example of this is Facebook’s News Feed, which changes according to the user’s personal interactions with other users. If a user frequently tags a friend in photos, writes on his wall or “likes” his links, the News Feed will show more of that friend’s activity in the user’s News Feed related to the algorithm’s criteria. Indeed, just last month the wellknown Scandinavian iGaming brand Betsson announced that their Big Data team in Malta had developed various AI algorithms to solve different business requirements. Some specifically were improving the customer experience and ability to detect fraud, automate payment processes and customer segmentation. In-play betting, which has experienced enormous growth among both operators and players, has seen numerous developments in the use of AI and machine-based algorithms. This makes absolute sense, when you think about the thousands’ of in-play betting markets offered, never mind cash-out and partial cash-out offers now available. These businesses need to adopt a lean approach and reduce operational costs regarding headcount within the trading and risk teams, and improve margins. If they didn’t use machine-based AI, they would have to hire teams in their hundreds to monitor the betting transaction flow, and likely it would be an unviable product as they would not be able to provide the volume and depth of in-play markets currently available to bet on. In user journeys and user experience or the actual digital sales funnel, programmatic advertising and real-time buying based on hard data or data science are using more machine-based algorithms to drive and expand new player acquisition programs. In fact there is now an application called the Grid.io, developed by a US software digital company in San Francisco, which uses AI to design and optimize websites in real-time and by device. Wave goodbye to front-end developers for desktop, tablet and mobile!
Certainly, everything and everyone has the potential to be distilled down to a machine-based algorithm. Just as couples will be able to determine the sex and genetic disposition of their child, so shall business determine processes and people. Machine-based learning is however only as good as the human programming the initial algorithms, and that code needs continual checking and for a human to actually make reasoned judgement on its validity. Or will it be the case that another AI code will be charged with checking the other code, ad infinitum? Ultimately, the future of AI application is within our hands as the creator. We will need to choose this future wisely.
“Just last month the well-known Scandinavian iGaming brand Betsson announced that their Big Data team in Malta had developed various AI algorithms to solve different business requirements.”
“Grid.io, developed by a US software digital company in San Francisco, uses AI to design and optimize websites in real-time and by device. Wave goodbye to front-end developers for desktop, tablet and mobile!”