EU delays AI Act enforcement by 16 months after industry backlash, adds deepfake ban

EU hits snooze on AI Act rules after industry backlash

Brussels says it's simplification, critics may call it retreat Europe's much-hyped AI rulebook just hit the regulatory equivalent of "snooze for 16 months" after Brussels quietly caved to industry pressure and agreed to simplify and delay key parts of the AI Act. European Union lawmakers and member states reached a provisional agreement Thursday on the so-called "Digital Omnibus on AI," a cleanup package that trims parts of the bloc's flagship AI law after months of complaints from industry that the rules were becoming unworkable. The headline change pushes back enforcement of rules covering systems in so-called high-risk AI areas such as biometrics, critical infrastructure, education, employment, migration, and border control until December 2, 2027, a 16-month delay from the previous August 2, 2026, deadline. AI systems embedded in products such as lifts and toys are now getting even longer, with compliance deadlines stretching to August 2, 2028. This marks a big win for tech firms and industry groups that have spent months pressuring the European Commission (EC) to soften the AI Act. Earlier this week, executives from companies including ASML, Airbus, Ericsson, Nokia, SAP, Siemens, and Mistral AI publicly warned that Europe risked over-regulating itself out of the global AI race. Officials maintain the delay is about timing, not watering down the law. The EC argues the rules are moving faster than the standards needed to support them, and companies still lack the guidance and technical tools required for compliance. The deal marks a notable rollback in the EU's digital rulebook after years of Brussels proudly marketing itself as the world's tech cop. That has come under mounting pressure from both Washington and European industry, which increasingly argues the bloc has become very good at regulating technologies it struggles to produce at scale. The revised package gives smaller companies a bit more breathing room and tries to untangle parts of the AI Act from existing product safety laws. Much of the focus is on industrial sectors where manufacturers had warned they were facing overlapping requirements and duplicate compliance work. Not every change loosened the law. The agreement adds a ban on AI systems used to create non-consensual sexual deepfakes and child sexual abuse material following global backlash over abusive uses of generative AI tools, including xAI's Grok chatbot. Providers claiming exemptions from high-risk classification will also still need to register those systems in the EU database. Commission President Ursula von der Leyen welcomed the agreement in a post on X, saying it would provide "a simple, innovation-friendly environment" while strengthening protections for citizens. Henna Virkkunen, the Commission's Executive Vice-President for Tech Sovereignty, Security and Democracy, tried to split the difference between deregulation and reassurance. "Our businesses and citizens want two things from AI rules," she said. "They want to be able to innovate and feel safe." For the time being, it seems Europe has decided that keeping its AI industry alive matters slightly more than making it fill out another stack of forms before 2027. ®

EU countries, lawmakers clinch provisional deal on watered-down AI rules

BRUSSELS, May 7 (Reuters) - EU countries and European Parliament lawmakers on Thursday agreed to watered-down landmark artificial intelligence rules, including delaying their implementation, in a move which critics say shows Europe caving to Big Tech. The tentative agreement, which needs to be formally endorsed by EU governments and the European Parliament in the coming months, came after nine hours of negotiations. "Today's agreement on the AI Act significantly supports our companies by reducing recurring administrative costs," Marilena Raouna, Cyprus's deputy minister for European affairs, said in a statement. Cyprus currently holds the rotating EU Council ⁠presidency. The changes to the AI Act, which entered into force in August 2024 with key elements to be enforced in stages, are part of the European Commission's push to simplify a slew of new digital rules. The simplification drive came after businesses complained about overlapping regulations and red tape that hamper their ability to compete with U.S. and Asian rivals. EU governments and lawmakers agreed to delay rules on high-risk AI systems such as those involving biometrics or related to critical infrastructure and law enforcement to December 2, 2027 from a previous deadline of August 2 this year. They also ⁠agreed to exclude machinery from the AI Act as it is already subject to sectoral rules, ceding to pressure from businesses. There was also agreement on a ban on AI practices which create unauthorised sexually explicit images, a move responding to such content generated by Elon Musk's xAI chatbot Grok on X and ⁠sexually intimate deepfakes produced by Grok. The ban will apply from December 2. "By the end of this year everyone, but especially women and girls will be safe from horrific nudifier apps being widely available on ⁠the EU market. Today we put a clear end to this kind of violence against people and children," said Dutch lawmaker Kim van Sparrentak. Mandatory watermarking of AI generated output will apply ⁠from December 2. The AI rules, which were triggered by concerns about the impact of the technology on children, workers, companies and cybersecurity, are still considered the strictest in the world even after the changes. Reporting by Foo Yun Chee; Editing by Jacqueline Wong and Lincoln Feast. Our Standards: The Thomson Reuters Trust Principles., opens new tab * Suggested Topics: * World * Data Privacy * Public Health Foo Yun Chee Thomson Reuters An agenda-setting and market-moving journalist, Foo Yun Chee is a 21-year veteran at Reuters. Her stories on high profile mergers have pushed up the European telecoms index, lifted companies' shares and helped investors decide on their next move. Her knowledge and experience of European antitrust laws and developments helped her break stories on Microsoft, Google, Amazon, Meta and Apple, numerous market-moving mergers and antitrust investigations. She has previously reported on Greek politics and companies, when Greece's entry into the eurozone meant it punched above its weight on the international stage, as well as on Dutch corporate giants and the quirks of Dutch society and culture that never fail to charm readers.

Brussels strikes deal to thin out AI Act and outlaw nudification apps

After two failed trilogues, Parliament and Council finally landed a compromise that pushes the high-risk compliance deadline to December 2027, lightens paperwork for smaller firms, and writes a long-promised ban on non-consensual intimate imagery into Europe's flagship AI law. The European Commission confirmed on Wednesday that negotiators from the Parliament and the Council had finally reached political agreement on the so-called AI Omnibus, the package of amendments designed to soften the application of the bloc's flagship Artificial Intelligence Act and bolt on a ban on AI-generated non-consensual intimate imagery. It took three rounds to get there. the failed 28 April session collapsed after roughly twelve hours of haggling over how AI built into regulated products should be assessed for conformity. A Wednesday session, scheduled at short notice ahead of a 13 May fallback date, closed the gap. Executive vice-president for tech sovereignty Henna Virkkunen, who pushed the simplification drive through the College of Commissioners last November, said the deal would let companies "focus on building, not on paperwork", framing it as proof that Europe can keep its rules-based approach while making them workable for industry. The headline change is the timeline. Obligations on standalone high-risk AI systems listed in Annex III, covering biometrics, education, employment, essential services, law enforcement, justice and border management, will now apply from 2 December 2027 rather than 2 August 2026. Rules for AI embedded in regulated products under Annex I take effect on 2 August 2028. For companies sitting on partly built compliance programmes, that buys roughly sixteen extra months. Brussels insists the postponement is a function of unfinished standards work, not a retreat: harmonised standards from CEN-CENELEC and a fuller library of guidance documents are the precondition for switching the obligations on. Smaller firms get more concrete relief. The agreement extends a set of simplifications already available to SMEs to small mid-cap companies, including templated technical documentation, lower fees and easier access to regulatory sandboxes. The intent, repeated through the Commission's press release, is to scale obligations to organisational size rather than apply a single compliance model to every provider in the value chain. The most politically charged element is the new prohibition on AI systems that generate child sexual abuse material or that produce non-consensual intimate images of identifiable people. Lawmakers had been pushing for it since the late-2025 controversy over Grok's nudification scandal, and Parliament made it a red line for the trilogue. The text now bans the placing on the market and use of AI tools whose primary purpose is to undress people in images or to depict identifiable individuals in sexually explicit scenarios without consent. Companies have until 2 December 2026 to bring existing products into line. The prohibition does not apply where developers have implemented effective safety measures to prevent generation and misuse, a carve-out negotiated to spare general-purpose models that already filter such outputs. TNW reported on the political agreement on intimate deepfakes when Parliament locked it into its mandate in late March; the trilogue text largely tracks that position, though enforcement now sits squarely with national market-surveillance authorities and the AI Office rather than with sectoral regulators. Critics will note that the package leaves the AI Act's core architecture intact. The risk-based pyramid stays. Foundation-model rules, in force since August 2025, are untouched. The Code of Practice for general-purpose AI providers continues to apply on a voluntary basis. Watermarking obligations on AI-generated content slip from February to December 2026 but remain mandatory. Civil-society groups, more than forty of which signed a letter against the Omnibus in April, have argued the simplification narrative obscures real cuts in fundamental-rights protection, particularly around biometric identification and AI in schools. Their concerns survive the deal: the trilogue did not reopen the substantive obligations, only their timing and paperwork. Industry, by contrast, has read the package as part of a broader competitiveness drive that includes the GDPR simplification and the Data Act review. The agreement bears that out: every concession in the AI Omnibus is procedural rather than substantive. The political agreement still needs formal endorsement by the Parliament's plenary and by ministers in the Council, expected before the summer recess. Without that, the original 2 August 2026 high-risk deadline applies, a scenario the Commission has spent six months trying to avoid. National authorities, meanwhile, get a parallel job: the simplified documentation forms, sandbox templates and SMC guidance need to be in place well before the new deadlines, or the relief on paper will not translate into relief in practice.

EU simplifies rules for AI companies in tentative deal

EU states and the Parliament reached an agreement to simplify the AI laws for businesses. Here's a look at what has changed. Member states of the European Union and the European Parliament reached a provisional deal to simplify the bloc's artificial intelligence (AI) rules as part of an omnibus package. The tentative agreement includes a delay for some major obligations for AI systems to prevent legal and commercial uncertainty, according to European Parliament officials. The Digital Omnibus on AI was proposed five months ago as a way for Europe to boost competition in the AI sector by simplifying compliance with the EU Artificial Intelligence Act. However, the proposed legislation has been controversial, with some accusing the EU of watering down its rules for AI systems. "We are not weakening any safety rules; we are clarifying the rules for companies in Europe," said Arba Kokalari, the rapporteur for the European Parliament's Internal Market committee. "The current state is that companies are confused about whether they should follow the AI act or sectoral legislation ... companies should not be regulated twice for one thing." Euronews Next takes a closer look at what was approved or changed. The EU AI Act classifies AI systems according to the level of risk they pose to health, safety, and fundamental rights, ranging from minimal risk to unacceptable risk, with stricter obligations imposed as the risk level increases. High-risk AI systems are those that will be embedded in critical infrastructure, education, employment, migration, asylum and border control. This category previously included products classified as "machinery," such as smart home appliances. Thursday's agreement gives so-called "high-risk" AI systems an extra year, until December 2027, to comply with EU legislation. The deadline extends to August 2, 2028, for AI used in products such as lifts or toys. There are also simpler rules for small and medium-sized businesses to "avoid duplication between sectoral and AI rules," the commission said in a press release. Companies developing AI systems also get access to a "EU-level sandbox" that will let them test their products before they enter the market The Digital Omnibus also prohibits AI systems that generate non-consensual sexually explicit and child sexual content, such as AI "nudification apps" that digitally remove people's clothing that strip subjects of their clothes. The ban covers explicit images, videos or audio created without a person's consent. Companies will have until December 2 to bring their systems in line with the EU's new regulations and to put in place mandatory watermarking of AI-generated content. Renew Europe lawmaker Michael McNamara told reporters that the new rules will apply to any photos where the person's "intimate parts" are exposed. when Elon Musk's AI chatbot Grok was used to generate sexually explicit AI images of women and children online. "We wanted to have clarity on what we think about [nudification apps] in Europe and that we are not accepting of it," said Arba Kokalari, the rapporteur for the European Parliament's Internal Market committee. The rules only apply to content that clearly depicts a human being, not a synthetic AI character, according to McNamara. The agreement still requires formal approval from the European Parliament and EU member states.

EU countries, lawmakers clinch provisional deal on watered-down AI rules - The Economic Times

EU countries and European Parliament lawmakers have agreed to a watered-down AI Act, delaying implementation for high-risk systems until December 2027. This move, aimed at reducing administrative costs and easing business competition, includes a ban on AI-generated explicit images and mandatory watermarking for AI output.EU countries and European Parliament lawmakers on Thursday agreed to watered-down landmark artificial intelligence rules, including delaying their implementation, in a move which critics say shows Europe caving to Big Tech. The tentative agreement, which needs to be formally endorsed by EU governments and the European Parliament in the coming months, came after nine hours of negotiations. "Today's agreement on the ⁠AI ⁠Act significantly supports our companies by reducing recurring administrative costs," Marilena Raouna, Cyprus's deputy minister for European affairs, said in a statement. Cyprus currently holds the rotating EU Council presidency. The changes to the AI Act, which entered into force in August 2024 with key elements to be enforced in stages, are part of the European Commission's push to simplify a slew of new digital rules. The simplification drive came after ⁠businesses complained about overlapping regulations and red tape that hamper their ability to compete with U.S. and Asian rivals. EU governments and lawmakers agreed to delay ⁠rules on high-risk AI systems such as those involving biometrics or related to critical infrastructure and law enforcement to December 2, 2027 from a previous deadline of August 2 this year. They also agreed to exclude machinery from the AI Act as it is already subject to sectoral rules, ceding to pressure from businesses. There was also agreement on a ban on AI practices which create unauthorised sexually explicit images, a move responding to such content generated by Elon Musk's xAI chatbot Grok on X and sexually intimate deepfakes produced by Grok. The ban will apply from December 2. "By the end of this year everyone, ⁠but especially women and girls will be safe from horrific nudifier apps being widely available on the EU market. Today we put a clear end to this kind of violence against people and children," said Dutch lawmaker Kim van Sparrentak. Mandatory watermarking of AI-generated output will apply from December 2. The AI rules, which were triggered by concerns about the impact of the technology on children, workers, companies and cybersecurity, are still considered the strictest in the world even after the changes.

EU countries, lawmakers clinch provisional deal on watered-down AI rules

BRUSSELS, May 7 (Reuters) - EU countries and European Parliament lawmakers on Thursday agreed to watered-down landmark artificial intelligence rules, including delaying their implementation, in a move which critics say shows Europe caving to Big Tech. The tentative agreement, which needs to be formally endorsed by EU governments and the European Parliament in the coming months, came after nine hours of negotiations. "Today's agreement on the AI Act significantly supports our companies by reducing recurring administrative costs," Marilena Raouna, Cyprus's deputy minister for European affairs, said in a statement. Cyprus currently holds the rotating EU Council presidency. The changes to the AI Act, which entered into force in August 2024 with key elements to be enforced in stages, are part of the European Commission's push to simplify a slew of new digital rules. The simplification drive came after businesses complained about overlapping regulations and red tape that hamper their ability to compete with U.S. and Asian rivals. EU governments and lawmakers agreed to delay rules on high-risk AI systems such as those involving biometrics or related to critical infrastructure and law enforcement to December 2, 2027 from a previous deadline of August 2 this year. They also agreed to exclude machinery from the AI Act as it is already subject to sectoral rules, ceding to pressure from businesses. There was also agreement on a ban on AI practices which create unauthorised sexually explicit images, a move responding to such content generated by Elon Musk's xAI chatbot Grok on X and sexually intimate deepfakes produced by Grok. The ban will apply from December 2. "By the end of this year everyone, but especially women and girls will be safe from horrific nudifier apps being widely available on the EU market. Today we put a clear end to this kind of violence against people and children," said Dutch lawmaker Kim van Sparrentak. Mandatory watermarking of AI generated output will apply from December 2. The AI rules, which were triggered by concerns about the impact of the technology on children, workers, companies and cybersecurity, are still considered the strictest in the world even after the changes. (Reporting by Foo Yun Chee; Editing by Jacqueline Wong and Lincoln Feast.)