“There is room for everyone on Earth: the entire human family must find the necessary resources to live properly thanks to nature itself, God’s gift to its children, and through the effort of its work and his creativity” (Benedict XVI, Caritas in veritate, 50).

Another phenomenon, more profound than freelancing, has today come to strongly challenge the notion of human work and worries, according to all studies, a significant part of the population: artificial intelligence. This innovation takes us back to the anthropological question that we asked at the beginning of this book: what is man?

Many books and articles explain very well what the term “artificial intelligence” means scientifically. So I’m not going to treat it in too technical a way. Born in the 1950s, it can be summed up as follows, in the words of John McCarthy (considered the father of AI): “The science and engineering to make intelligent machines.” So for decades we have been faced with “intelligent” algorithms to which we have given rules in order to then ask them to solve problems. But this was very complex and limited.

In recent years, giant strides have been made with so-called “machine learning,” a subset of artificial intelligence developed in the 1980s. Machine learning equips machines with a learning capability that ranges from supervised learning (the typical example is the myriad of cat photos associated with the word cat) to the more recent “deep learning”. Both the computing power of machines and the mass of usable data (which we ourselves provide for free on social networks) have made it possible to drastically accelerate scientific research. Today, the searches we carry out on Google or the recommendations made to us on e-commerce sites are already applied artificial intelligence.

Deep learning, which is therefore a subset of the subset, attempts to bring the functioning of the machine closer to that of the human brain. A child does not need to view a million photos of cats to affirm that what he or she sees is a cat; he experiments for himself. While the understanding of the brain and neural systems has progressed enormously in recent years, deep learning is trying (it is still far) to imitate neurobiological processes.

It works on several levels and is based on artificial neural networks which allow the machine, without being guided, to define what it sees. Here again, there are several subsets. What we call “reinforcement learning” is, for example, DeepMind (Google) which, before beating the world champion in the game of Go, analyzed numerous games, but also played millions against himself; so he taught himself. And recently, researchers are working on unsupervised learning, for which the machine builds an algorithm from its observation alone.

We are still in the early stages, artificial intelligence is very far from understanding human reasoning, as recalled in the Al Index report published in November 2017, although many fears come from there. Without going into too many technical details, it is important to clarify two things at this stage: the first is that the performance achieved by deep learning is impressive; the second is that we don’t really understand why, that is to say that there is an operation and results that escape even the designer.

While robotization is already well present in our factories, weak artificial intelligence, the one we currently know, therefore already seems to be able to acquire human skills: translate a discussion, label images, detect fraud, carry out a medical diagnosis. .. As a result, many low cognitive skill jobs appear to be under threat, and this disturbs us. Will there still be human work tomorrow? Accountants, drivers, traders and even surgeons are growing gray hair…

According to Jacques Bughin and Éric Hazan’ of McKinsey, three factors tend to confirm this deployment of artificial intelligence. First, venture capital funds and other financial investors have tripled their investments in these areas over the past three years’; then, the large technological groups have also invested considerable sums; finally, their study of three thousand companies shows that two thirds of them are open to integrating artificial intelligence into their organization.

Furthermore, we can note, perhaps with some concern, the speed with which China has taken the A turn by making a phenomenal mass of usable data available to companies; in three years, Chinese technological progress has been considerable.

In 2013, researchers at the University of Oxford came to the famous conclusion that 47% of jobs in the United States were threatened by artificial intelligence and robotics. This conclusion is today undermined by all those who consider, on the one hand, that man has a relational capacity and a creative talent which remain essential in many technical works and, on the other hand, that artificial intelligence will open the way to new human roles.

Mark Zuckerberg himself nevertheless expressed his concern during a speech he gave at Harvard in 2017:

When our parents graduated, their purpose in life came, simply, from their job, from their church, from their community. But today, technology and automation are destroying many jobs. The number of members is decreasing in all communities. Many people feel disconnected and depressed and are trying to fill a void. […] I met workers who know that their jobs will no longer exist and who are trying to find their place.

This text is an extract from the book “GOD, THE COMPANY, GOOGLE AND ME” written by Thomas JAUFFRET.

We invite you to read the following article “The company becomes the arbiter of public morality“.