Objectives
- Understand what artificial intelligence really is
- Distinguish AI from traditional computing
- Identify concrete applications of AI in our daily lives
- Grasp the issues and prospects of this technology
Artificial intelligence now occupies a growing place in our society. Present in many of the tools and services we use every day, it has become one of the most striking and most discussed topics of our time.
The volume of data circulating around the world is considerable: in just a few seconds, millions of pieces of information are exchanged. Tens of millions of e-mails are sent, several million instant messages circulate, and countless pieces of content are published on digital networks. At the same time, millions of automatic translations are carried out every day and new data is generated constantly.
For a long time, artificial intelligence remained confined to research laboratories and universities. Its concrete applications in businesses and among the general public remained relatively limited.
The emergence of Big Data and the considerable increase in computing power have profoundly changed this situation. Today, AI is used in many sectors and is helping to transform numerous professional and economic activities.
Defining artificial intelligence precisely is not a simple task. This difficulty is largely due to the fact that human intelligence itself remains a complex concept to characterise.
The main difference between artificial intelligence and traditional computing lies in the way problems are solved.
Classical computing often relies on deterministic algorithms capable of systematically examining a very large number of possibilities in order to identify the best solution.
Advantage: This approach can be extremely efficient.
Limitation: It sometimes becomes impractical when the number of combinations to analyze is too large.
AI takes a different approach. Rather than systematically exploring all possible solutions, it quickly seeks out relevant solutions by relying on learning, optimization or reasoning mechanisms.
Advantage: It makes it possible to obtain results within reasonable timeframes, even for complex problems.
Limitation: It is not always possible to guarantee that the solution found is the best among all those that exist.
(Exhaustive exploration) Best solution
(long time)
(Learning & optimization) Relevant solution
(short time)
When it first emerged, artificial intelligence pursued an ambitious goal: to replicate the workings of the human brain and create machines capable of thinking like a human being. Even though this goal has not yet been fully achieved, AI has already proven its effectiveness in many fields.
- Medical data analysis — clinical diagnostics and predictions
- Machine translation — converting text between languages
- Image recognition — visual identification and classification
- Decision-making support — recommendations and analyses
- Task automation — execution of repetitive operations
- Improving the user experience — personalisation and adaptation of digital services
The rapid development of artificial intelligence sparks as much enthusiasm as it does questions. Its potential impact on several key areas remains at the heart of current debates.
Understanding the principles of AI is therefore essential to better grasp the opportunities it offers, but also the challenges it poses to society tomorrow.
To remember
- The artificial intelligence is a set of methods and technologies aimed at reproducing human abilities such as learning, reasoning and decision-making.
- AI stands out from traditional computing through its approach: it looks for relevant solutions quickly rather than exhaustively exploring all possibilities.
- The rise of AI was made possible by the emergence of Big Data and the considerable increase in computing power.
- AI has concrete applications in many fields: medicine, translation, image recognition, decision support and improvement of digital services.
- The rapid development of AI raises important issues concerning employment, privacy, security and the responsibilities linked to automated decision-making.