Observed Signal · May 7, 2026 · Regulation · Source: t3n · Impact: 4/5 · Sentiment: Neutral
EU agrees ban on sexualized deepfakes from Dec 2026
EU member-state representatives and the European Parliament have reached an agreement to amend the EU AI Act to ban AI applications used to create sexualized deepfakes, including depictions of child sexual abuse. The provisional deal must still be confirmed by the full Parliament and the Council, a step usually seen as formal. If adopted, enforcement would be led by the EU AI Office (KI-Amt der EU) starting 2 December 2026. The agreement also delays and phases other AI rules: mandatory watermarking and clear labelling of generated images and videos will be enforced from December 2026, while additional provisions will be phased in by December 2027. The move follows public debates triggered by incidents involving chatbots and high-profile national discussions on digital sexual violence.
An EU-level change to the AI Act affects AI providers, platform moderation, content labelling obligations and compliance timelines across the European market; enforcement by the EU AI Office with concrete deadlines (Dec 2026/2027) will require operational and product changes for major AI and platform companies.
Track TargetVideo Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your free Explorer workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- EU representatives and the European Parliament reached a provisional agreement to amend the EU AI Act to ban AI tools for creating sexualized deepfakes.
- The ban explicitly covers content that depicts sexual child abuse.
- The EU AI Office (KI-Amt der EU) would enforce the ban from 2 December 2026 if the reform is confirmed.
- Providers must label AI-generated images and videos with watermarks starting no later than December 2026; other enforcement actions are phased until December 2027.
- The agreement still requires formal confirmation by the plenary European Parliament and the Council of the EU member-state governments.
Connected Companies & Entities
4 Entities mappedOntology Mapping & Concepts
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
EU moves to ban nudifier deepfake apps
The EU Parliament has voted to ban so-called "nudifier" applications that use AI to generate or alter images and videos to depict sexually explicit activity or intimate body parts of an identifiable person without consent. The move responds in part to a recent "Grok" scandal and public outrage around AI-generated sexual imagery, including protests in Germany after prominent cases such as Collien Fernandes, who has reported fake nude images circulated under her name. Separately, Germany’s federal justice minister Stefanie Hubig plans a national draft law to criminalize production of pornographic deepfakes, allow victims to request account takedowns and permit short-term IP-address retention. Experts cited in the article warn of severe psychological harm to victims of AI sexualized imagery.
EU begins enforcing AI Act, tightens transparency rules
The European Commission will begin enforcing the EU AI Act on 2 August 2026 and has introduced stricter transparency requirements for AI use. Users must be able to recognize interactions with AI and identify AI‑generated or altered content (including clearly labeled deepfakes). Providers of general-purpose and high-capability AI models face additional obligations: risk assessments, security measures, documentation of training data and model architectures, and publication of summaries and copyright policies. Certain manipulative or harmful AI uses will be banned. Enforcement will be coordinated by a newly created EU Office for Artificial Intelligence together with national authorities (e.g., Bundesnetzagentur) and the European Data Protection Supervisor, supported by a scientific expert group and new reporting mechanisms. The law will be rolled out further through 2028 with additional high-risk rules phased in.
Swiss Parliament Seeks to Regulate AI Deepfake Providers
The Swiss National Council (Nationalrat) unanimously adopted a motion by Raphaël Mahaim (Green Party, Vaud) asking the government to make generative AI providers responsible for protecting against sexualized deepfake content and to establish a review body able to impose sanctions. The Federal Council supports the motion and plans to consult on ratifying the Council of Europe’s AI conventions; the upper house (Ständerat) must still decide. The motion cites examples involving the AI tool Grok on platform X. Separately, the European Parliament approved an amendment to the EU AI Act that would ban AI applications able to produce sexualized deepfakes, with 423 votes in favour; formal approval by EU member states is still required and enforcement by the EU AI Office would begin on 2026-12-02. The draft law also requires clearer definitions of sexualized content and mandates marking/guidance such as watermarking for generated media.
Track Real-Time Market Signals & Shifts
Set up custom watchlists to receive automated, evidence-grounded executive digests whenever material signals or shifts occur across your tracked landscape.