The European Union's Artificial Intelligence Act is the most significant project in the field of AI regulation. This is probably due to the fact that it constitutes the first binding legal project. It is less than two years since it has generated an intensive discussion, first among the European institutions, but also between the various actors involved in AI: scientific and academic associations, AI developers, GAFAM, and Human Rights NGOs. This paper seeks to analyse some of the main issues raised by the AI Act: why is there a need for a European regulation? What exactly is the AI Act and what are its major challenges?