The dictionary of the Royal Spanish Academy defines “intuition” as “the faculty of understanding things instantaneously, without the need for reasoning”, and also as an “intimate and instantaneous perception of an idea or a truth that appears as evident to the one who has it”. And this is an extremely human quality that now reaches the world of machines. Indeed, at the intersection of these definitions, lies what is considered to be the fourth generation of artificial intelligence (AI): artificial intuition.

This is a paradigm that seeks to enable computers, even when they do not receive any specific order or data input, to identify risks and opportunities in a given scenario.

Thus, it makes it possible to detect threats, problems and opportunities without knowing in advance what exactly is being sought. In a way, it “emulates” human intuition, in particular the most powerful and inscrutable part of the brain: common sense.

For unknown scenarios

Algorithms today depend to a large extent on the data ingested into them. Although the concept of “synthetic data” has been consolidated for contexts in which there is no historical or previous information, this situation continues to encounter limits, especially when it is necessary to face unknown or unprecedented scenarios.

This is where the need for “instinct” to play a key role in decision-making arises. As if it were a detective trying to find unconnected pieces to reach the answer to a problem that is nothing like any other.

This new paradigm takes the available data set (which does not need to be large in volume, as is generally the case in AI and machine learning applications) and finds correlations or anomalies.

It uses qualitative and mathematical models to determine a big picture, looks for an initial solution, and then goes back to the beginning to fill in gaps and find patterns to “sense” what is needed and, as a consequence, clarify the solution. No matter how subtle the indicators, nothing will go unnoticed.

In this way, you can find something that is not working as it should, anticipate a problem before even the first symptoms are present, or find an opportunity for the company to use even when the signs are insufficient.

Some use cases

We said that artificial intuition is the fourth generation of AI. The historical path would be configured in the following way: the first is descriptive analytics (analyzes what has happened), the second is diagnostic (finds causal factors) and the third is predictive (evaluates previous scenarios to understand what will happen).

Prominent use cases for artificial intuition include the analysis of new customer behavior within a retail store or the generation of personalized offers for a specific consumer at a particular time. In a more evolved scenario, it could have the ability to detect future consumer needs that have not yet explicitly manifested themselves, giving retailers the ability to offer them in advance while gaining a huge competitive advantage.

It is also useful for improving the ability to detect financial fraud and fakes, which makes it a very useful tool for finance or security companies.

Security by hunch

On the other hand, it implies a remarkable evolution in the field of cybersecurity, as it significantly decreases the number of false positives in analysis, which in turn leads to one of the major problems in the industry: alert fatigue, which is when those responsible ignore some warnings due to the volume they receive.

Similarly, it improves accuracy in relation to what is known as “false negative” since, as mentioned, it detects anomalies even when they are extremely subtle or imperceptible, two of the great qualities that attackers generally have. It might even be able to find within what at first glance is an unimportant transaction, a new and previously undetected pattern of a new cyberattack format.

Artificial intuition is a technology that is just taking its first steps. However, given the features it offers and the benefits it promises, it could become more successful in the not-so-distant future.