Petra Molnar, human rights lawyer, speaks about the dangers of high-risk AI systems and state efforts to sabotage its regulation, as seen in the recent watering down of the European Union’s AI Act.