The European Parliament wants to divide content generated by artificial intelligence into different risk categories. While some systems could be banned, services like the popular ChatGPT would hardly be affected.