It will not be an exaggeration to say that one of the most misused and misinterpreted words — in common as well as in business context — is AI. There is a quote by Cassie Kozyrkov that gets it exactly right:

If it is written in Python, it's probably machine learning. If it is written in PowerPoint, it's probably AI.

In today's tech world, using the term artificial intelligence or A.I. has become increasingly fashionable for any piece of software that uses a simple neural network — typically a pattern recognition system just capable enough to sort certain things into categories. There are even more shameful cases of decision-tree-based systems being marketed under the guise of AI. AI has also become a fashion item for corporate strategy. Bloomberg Intelligence's Michael McDonough tracks mentions of "artificial intelligence" in earnings-call transcripts, and the uptick is striking. Jerry Kaplan put it bluntly: "Artificial intelligence, it seems, has a PR problem…"

Understanding AI

Computer science defines AI research as the study of "intelligent agents": any device that perceives its environment and takes actions that maximise its chance of successfully achieving its goals. A more elaborate definition characterises AI as "a system's ability to correctly interpret external data, to learn from such data, and to use those learnings to achieve specific goals and tasks through flexible adaptation."

There are three types of AI

  1. Narrow / weak AI. Face and speech recognition, voice assistants, driving a car. Apple's Siri, IBM's Watson, and most "AI" features in consumer products. This is the only category we have plausibly achieved.
  2. General / strong AI. Machines that mimic human intelligence — learning, reasoning, problem-solving across domains. The very lack of understanding of the human brain itself is the biggest deterrent; modern science is still far from this.
  3. Artificial superintelligence. A hypothetical AI that surpasses human intelligence and behaviour — where machines become self-aware. Matrix.

Scientism around AI

The usage of these buzzwords — and the scientism that comes with them — must be stopped. For businesses, the use of the term must be well-thought and well-defined in the context of goals and data strategy. It is a sure sign of poorly defined goals when these terms appear more often in meetings and presentations than in actual systems.

The story doesn't end at AI. Big data and analytics have become similarly ubiquitous on PowerPoint slides — but that's the subject of another article.