Algorithms are often disparaged, but they are also sometimes overrated. To understand their place in our world it is also important to understand their limits.

In the age of algorithms, new fears have arisen, including the fear that one day we will be overtaken, or even enslaved, by new beings, transhumans favored by natural selection, but also, more simply, by computers or algorithms more intelligent than us. This leads us to a question that computer scientists have been asking since the 1950s: Can an algorithm be intelligent? This question brings up two others. What does the adjective intelligent mean? Can we create an intelligent being?

Algorithms have become an essential component of our professional lives and social interactions, in health care, transportation, commerce, industry. Algorithms are transforming the natural sciences, social sciences, and the humanities, and in doing so, enrich our knowledge. They allow technology to continually push beyond the limits of the possible.

We used to believe that intelligence, like speech, culture, and self-awareness, made us unique. However, diluting intelligence across a variety of faculties contributes to blurring the boundary that separates man and machine. Man is better at speaking Japanese, but machines are better than man at playing chess. Perhaps one day machines will also be better at speaking Japanese. The difference between man and machine seems to be more a matter of degree than of nature, a distinction that enabled us to conceive of the idea of augmented man.

What could we still say was unique about us in the twentieth century?

Algorithms transform the relationship between human beings and nature, and in doing so, transform nature itself. This leads us to examine the relationship between the digital revolution and another transformative factor in our world today: the ecological transition.

The community is usually defined socially as a group of human beings whose life together is made possible by respecting a certain number of rules that define the rights and obligations of each of them. According to this definition, the members of the community are women and men. However, a gradual evolution of this concept has led us to consider that groups of human beings, for example, companies, and associations, can also have rights and obligations and, therefore, can also be considered members of the community.

In using the example of the professions of driver and translator, we implicitly assumed that drivers and translators would always exist. However, it is also possible that these professions may one day disappear if, at some point, algorithms for driving a car or translating a text perform as well as, or even better, than a human. This is also true for many other professions. Of course, this transition also paves the way for new professions to design, implement, and accompany all of these algorithms, but in the age of algorithms, much less work may be required to provide the same goods or services as before.

We have been using symbolic algorithms since the advent of writing, five thousand years ago. How is it then that this concept has suddenly become such a hot topic in the public sphere today? To explain this, we need to look into objects other than algorithms – computers and programs.

In 2002, during a performance/installation, 35 Hours of Work, Benjamin Sabatier sharpened pencils seven hours a day for five days. Sharpening pencils in this way for thirty-five hours is a deviant act, because custom dictates that we use a pencil sharpener for a few seconds, to sharpen a pencil, after which we put it away in a drawer until we need it again. As a consequence, a pencil sharpener is used only a few minutes per year.

Algorithms and computers help with everything. But what tangible purpose do they serve? The remarkable diversity of their uses derives from their universality.

Why is the notion of privacy at the core of the issues emerging in the age of algorithms?

We have certain expectations of the algorithms we use. For example, we would like them to be fair. These properties are essential for the peaceful coexistence of humans and algorithms, and for establishing a climate of trust in the community. They are even more critical when these algorithms exercise a certain power, such as when an algorithm makes the decision to approve or refuse a bank loan. What exactly are these expectations?

A scientific revolution does not only create new knowledge. It also generates new ways of thinking, new ways of asking questions, and new ways of answering them. Before the scientific revolution at the beginning of the seventeenth century, when questions were raised, for example, about whether or not blood circulated in the body, people looked for answers in the ancient texts. Aristotle and Galen taught that blood did not circulate in the body; therefore, the question had been answered. How did Aristotle and Galen know what they knew? That question was not posed. They were more knowledgeable than us, and that was sufficient.

Can an algorithm do harm? Unfortunately, the answer to this question is fairly predictable. Like any tool, an algorithm can be used for better or for worse. For example, if most people receive a pay stub every month produced by an algorithm, it would be very easy to modify the algorithm to add a bonus for certain employees depending on their nationality, skin color, gender, or political opinions. Such discrimination is morally reprehensible whether it is carried out by a human or an algorithm.

Where does the apparently paradoxical idea of the economy of free come from?

In many countries, governments are now conscious of the importance, in the age of algorithms, of computer science education. In the United Kingdom, for example, the Royal Society published a report in 2012, “Shutdown or Restart: The Way Forward for Computing in UK Schools,” recommending the introduction of computer science education in schools. At about the same time, Google chairman Eric Schmidt’s devastating critique attracted a lot of attention: “I was flabbergasted to learn that today computer science isn’t even taught as standard in UK school. Your IT curriculum focuses on teaching how to use software, but gives no insight into how it’s made.” As a result, the United Kingdom engaged in a complete reconstruction of its computer science educational system. In the United States, an “advanced placement” course called AP Computer Science was introduced in 1984 as a specialized high school subject. But the education system is quite fragmented and schools are not required to offer it. Barack Obama declared in 2016, “In the new economy, computer science isn’t an optional skill. It’s a basic skill, right along with the three R’s.” In 2018, only about 135,000 out of the 15 million high school students in the United States were enrolled in the course.