What happened
On June 15, 2026, the European Parliament adopted in plenary session a set of simplification amendments to the EU AI Act (press release made available June 16, 2026). The amendments streamline compliance obligations for companies while preserving the law's core risk-based architecture. Crucially, the Parliament outright banned 'nudifier' AI applications — tools that generate non-consensual intimate imagery — as a prohibited AI practice. The vote confirms the August 2, 2026 applicability deadline for high-risk AI system obligations and sets December 2026 for watermarking/transparency enforcement to begin.
Why it matters
This is a binding legislative action by one arm of the EU co-legislature. The nudifier ban takes immediate effect as a prohibited practice under the AI Act's Article 5 framework. The simplified compliance pathway affects thousands of companies deploying high-risk AI in the EU, potentially reducing their documentation burden while bringing the August 2 deadline into sharp focus. Any AI system that generates synthetic imagery of real individuals will need to assess whether it falls within the banned category.
Action needed
Operators of AI systems that generate or manipulate intimate imagery must immediately assess prohibition exposure. All providers of high-risk AI systems must be ready to comply with the August 2, 2026 obligations. Review updated conformity assessment procedures under the simplification measures. Prepare for watermarking obligations entering enforcement by December 2026.