Artificial Intelligence: At the heart of a technology revolution
The rise of artificial intelligence
With the rise of artificial intelligence (AI), we are witnessing the beginnings of a major technology revolution. The revolution started five years ago with applications in image and voice recognition at companies such as Google, Microsoft and IBM. Since then, these new technologies are spreading rapidly and have already disrupted a number of fields. They enable complex activities, such as self-driving cars and medical diagnoses, and can replace administrative assistants with intelligent agendas. Artificial intelligence is no longer a curiosity of interest only to a few magazines that specialize in new technology, such as Wired and Motherboard, but rather a topical issue the impact of which is being felt in everyone’s lives (see Tout le monde en parle, Time, The Atlantic and Forbes). This technology revolution is disrupting our lives faster than experts in the field thought possible.
What is AI?
Artificial intelligence refers to the intelligence of machines and software. It involves concepts like reasoning, learning, perception and planning. Of the different facets of artificial intelligence, automating the learning process is what is creating the current revolution. Rather than designing software to recognize images or drive a car, the current approach is to program software that is capable of learning to recognize images or drive a car. In the jargon of the field, this is called machine learning.
How can a machine learn?
How software learns is not all that different from how humans learn.
For instance, to teach children what a car is, you show them several cars and tell them they are cars. After being exposed to a number of cars, they develop the concept of a car in their mind. Then they can use the concept to correctly identify future objects that are cars.
Computer learning works the same way. To help your computer learn to recognize cars, you have to show it many images and indicate which are cars. Using statistical algorithms, the software will learn and build what is called a predictive model. Once this model is developed, you can show it new images, and it will be able to recognize the images of cars among them.
The applications of AI
What form will artificial intelligence take in our lives? When we think of AI, we tend to imagine robots in humanoid form. In the movie Ex Machina (2015), for example, a billionaire in the technology industry tries to create an android that is indistinguishable from a human being. All signs point to the possibility that, relatively soon, technology will enable the creation of truly intelligent machines. However, for the time being, the AI revolution involves integrating ultra-specialized intelligence to existing products and services. So currently, AI is mainly being introduced into domestic objects, such as cars and robot vacuum cleaners. It is also used by Netflix, Amazon and Google to help them understand your needs and suggest products and services intelligently.
AI as an expert
These disruptions will not only influence domestic life, but will also revolutionize the working world. According to a study conducted at the University of Oxford, close to 47% of jobs are at risk of being replaced by artificial intelligence. The BBC has put a tool online that tells you how likely it is that your profession will one day be replaced by a machine.
Algorithms developed using AI techniques are already appearing in highly specialized fields that require complex cognitive tasks, such as law, medicine and finance. It is therefore possible to teach a machine how to do the research to identify the most relevant legal citations for a case, how to make a diagnosis to determine whether a tumour is benign or malignant and how to make the best investment decisions for a portfolio. Psychology won’t be left on the sidelines of this revolution. Researchers recently showed that it is possible for a machine to learn to recognize depressive traits by analyzing the way someone speaks.
In short, robotics revolutionized assembly line work, but in the coming years, artificial intelligence will no doubt revolutionize repetitive intellectual work.