The successful integration of AI into a business depends on the clarity of purpose, the quality of training data and having the knowledge and expertise to properly apply said data and train an algorithm appropriately.

This is why it takes a lot of work by skilled humans to make an AI that will help streamline processes and improve profitability, and an unusual example of this in action can be found in the unlikely world of chess.

Chess is one of the most popular games in the world, but whilst it was historically seen as a barometer for human intelligence, it is also a game that has no hidden information, no random elements and thus can be solved by a suitably powerful algorithm.

It was described as the “fruit fly of AI” for decades, as whilst it was a relatively simple application of AI compared to medical systems and other modern applications of AI, it could still be used to teach fundamental lessons about what AI can and cannot do.

At its most basic and least scientific level, Chess was perceived until the success of Deep Blue to be the ultimate showcase of intelligence. Both of its matches against grandmaster Garry Kasparov proved that in classical tournament conditions, AI could beat even the best player in the world.

There were caveats to this victory; the nature of the match format which allowed the IBM engineers to tweak the machine led to allegations of cheating, and the solvable nature of chess benefitted so-called brute force AI algorithms where powerful enough computers could go through millions of combinations to find the right move.

Regardless, it proved the principle that given enough processing power, even the most complex of human mind sports could be perfected by AI, and this approach would be applied not only to other games such as Go but to medical diagnoses and business functionality.