EU AI Act - Footnote 33

Note: The claim mentions 'Chatbots, deepfakes, emotion recognition systems' as examples under 'Limited' category, but the source only explicitly mentions 'emotion recognition systems' and 'deepfakes' in the context of transparency obligations. Chatbots are mentioned elsewhere as an example of AI systems. The claim states 'August 2, 2026' as the date when transparency obligations take effect, but the source states that the rules for high-risk AI systems will take effect on this date. The source does not explicitly state when transparency obligations take effect.