All our customers have tremendous interest in AI and ML – artificial intelligence and machine learning – across every industry and in companies of every size. Yet the very power of the phrase, and the breadth of interest, reveal one of its challenges: The term is too vague to be really useful for planning purposes. Is AI the same as ML? Most AI techniques are also machine learning techniques but many experts still differentiate between ML and AI – often based on the details of the approach taken.
To try and cut through some of this, we use this graphic to show the different ways AI and ML can play a role in operations. As always, we take a DecisionsFirst approach to this – our objective is to develop a decision service that automates all of part of your decision making accurately, consistently and legally while empowering the business to evolve it and manage it. In that context there are three roles for AI/ML:
Interface AI – the use of ML/AI to improve the interaction of the computer and its decision service with the outside (real) world.
Research AI – investigating data (structured or unstructured) to see what can be learned about the business or situation that generated that data.
Operational AI – ML and AI components that can be deployed into a decision service to make it more accurate, better able to consume and understand certain kinds of data and to replace or augment human decision-making.
I’ll talk more about each one in a subsequent post, starting with Interface AI.
