If artificial intelligence destroys work, it will only be the fault of big industries

About "digital economic revolution", the worldwide trend of the technological industry is to try to automate activities and remove human beings – too slow, expensive and unionized – from the production-distribution chain. And if on the one hand the policy must intervene to balance this trend, on the other hand we think of the current system to mess up the cards on the table, since when you need to solve software errors, you need (still) people in flesh and bone.

It happens with the apps. A case in point is Expensify, the app that allows you to automatically record expenses and reimbursements by taking a simple photo of the receipts. But if the software that analyzes the receipts goes into "botta", inevitably a human being is needed for this revision. With relative privacy issues because marketing and thousands of workers, including third parties, are involved. And that's what happens even with social media for checking questionable content. The algorithms that control them add up to both Facebook, YouTube is Twitter thousands of human moderators-overseers. IS Babel – «colossus» of professional translations – it has 50,000 reviewers between men and women to control and modify the translations produced by the artificial intelligence system for its customers. 50,000 reviewers able to read potentially confidential or strategic content.

And most of the time these human workers they are not simply filling software gaps, but they are also helping to form the learning component automatic services of these companies providing new examples on how to solve the problems of the technological system. Instead other times human beings are not used to interact with artificial intelligence systems, but work upstream, classifying necessary data for images, texts and audio.

Zooniverse is a "citizen science" web portal that allows users to participate in scientific research, from the classification of galaxies to the collection of climate data, to the reading of ancient papyrus; and serves to help academics analyze large data sets that are too complex for computers. Then there is MTurk (Mechanical Turk), Amazon's crowdsourcing marketplace that allows individuals and companies (known as "Requesters") to coordinate the use of human intelligence to perform tasks that computers are not currently capable of play.

This mixture of men and artificial intelligence essentially generates a privacy problem, also because the user-client is not always aware that some stranger is dealing with his data, maybe even very sensitive, such as those related to health. It is for example the case of the IBM Watson artificial intelligence system Health (specifically the IBM Watson for Oncology), which for its development – in some cases to validate the results – relies on over-qualified doctors, such as those at the Memorial Sloan Kettering Cancer Center in New York. Furthermore, there is the aspect that brings us back to the "digital economic revolution" and to the relative global trend of the technological industry that wants more and more automation and less human resources.

For now, to be honest, it's more like a perceived and overrated trend, but which serves the large industries of the sector to remove more and more bargaining power from human resources, making them believe that they are no longer needed or almost. Instead, at least until today, artificial intelligence is in many fields still substantially ineffective without human support. What does all this mean? That despite apocalyptic forecasts, artificial intelligence does not destroy jobs. But he will do it, because of (deliberate) the same artificial intelligence industry, because even now most of this work entrusted to the man is monotonous, poorly paid and deliberately isolated. Here it is then a good job for the future: the union of men against the artificial intelligence industry!