[{"data":1,"prerenderedAt":30},["ShallowReactive",2],{"nr-en-eu-ai-act-umsetzung-verzoegert-hochrisiko":3},{"slug":4,"title":5,"dek":6,"date":7,"time":8,"publishedAt":9,"updated":10,"updatedAt":10,"dateFmt":11,"updatedFmt":10,"kind":12,"tier":13,"author":14,"authorName":15,"topics":16,"tracker":22,"trackerLabel":23,"headlineStat":24,"image":25,"ogImage":26,"imageAlt":5,"csv":10,"minutes":27,"words":28,"html":29},"eu-ai-act-umsetzung-verzoegert-hochrisiko","EU AI Act Stalled: Implementation Deadlines for High-Risk AI Delayed","The EU has loosened its own implementation timelines for regulating dangerous AI systems. What's framed as relief for industry could give tech giants time to water down the rules.","2026-08-02","11:14","2026-08-02T11:14:00+02:00","","August 2, 2026","news","standard","ideal-syka","Ideal Syka",[17,18,19,20,21],"EU AI Act","AI regulation","Compliance","High-risk AI","Transparency requirements","\u002Feu-ai-act-fahrplan","EU AI Act Timeline","Fines up to €15 million from December 2026","\u002Fnewsroom\u002Fimg\u002Feu-ai-act-umsetzung-verzoegert-hochrisiko.webp","\u002Fog-nr\u002Feu-ai-act-umsetzung-verzoegert-hochrisiko.en.png",3,531,"\u003Cp>The EU has postponed implementation deadlines for core provisions of the AI Act – precisely the strictest rules governing high-risk AI systems. This shows that the world's first comprehensive AI rulebook, adopted in 2024, is already facing pressure before it's fully in force.\u003C\u002Fp>\n\u003Ch2>The essentials\u003C\u002Fh2>\n\u003Cul>\n\u003Cli>\u003Cstrong>From August 2026\u003C\u002Fstrong> major AI providers like \u003Cstrong>Google, Meta, and OpenAI\u003C\u002Fstrong> must label AI-generated content – with machine-readable digital watermarks and metadata\u003C\u002Fli>\n\u003Cli>\u003Cstrong>Fines from December 2026\u003C\u002Fstrong>: up to \u003Cstrong>€15 million or 3% of global annual revenue\u003C\u002Fstrong> for violations of labeling requirements\u003C\u002Fli>\n\u003Cli>\u003Cstrong>High-risk AI rules postponed\u003C\u002Fstrong> – original implementation deadlines loosened; official rationale: more time for EU and companies\u003C\u002Fli>\n\u003Cli>\u003Cstrong>From December 2026\u003C\u002Fstrong> the EU bans AI systems that generate intimate images without consent (nudify apps, deepfake pornography)\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch2>What changes concretely\u003C\u002Fh2>\n\u003Cp>From August 2026, first transparency obligations take effect: all AI applications in customer service – whether chat or phone – must be labeled. Synthetic voices, image and content generators too. Deepfakes must be visibly or audibly marked as manipulated, regardless of whether they contain political content or sexualized material.\u003C\u002Fp>\n\u003Cp>In December comes the ban on particularly problematic AI: systems that generate intimate images without consent are then prohibited in the EU. This affects nudify apps, deepfake pornography tools, and AI-generated child sexual abuse material. The EU is responding to incidents like features of the chatbot \u003Cstrong>Grok\u003C\u002Fstrong> from Elon Musk's firm \u003Cstrong>xAI\u003C\u002Fstrong>, which caused uproar in early 2026.\u003C\u002Fp>\n\u003Ch2>The core problem: high-risk rules delayed\u003C\u002Fh2>\n\u003Cp>The critical element is the postponement of high-risk rules. These were originally supposed to take effect sooner – they're the heart of the AI Act. The stricter the rules, the more dangerous the AI application is for people and fundamental rights. Yet the EU has loosened these timelines.\u003C\u002Fp>\n\u003Cp>Official justification: both the EU and companies need more time to implement. In reality, translating risk-based requirements into practice has proven more complex than initially thought. This is viewed both as necessary relief for industry – and as a critical delay by critics.\u003C\u002Fp>\n\u003Cdiv class=\"tbl-scroll\">\u003Ctable>\n\u003Cthead>\n\u003Ctr>\n\u003Cth>Measure\u003C\u002Fth>\n\u003Cth>Effective from\u003C\u002Fth>\n\u003Cth>Fines from\u003C\u002Fth>\n\u003C\u002Ftr>\n\u003C\u002Fthead>\n\u003Ctbody>\u003Ctr>\n\u003Ctd>AI labeling (texts, images, videos)\u003C\u002Ftd>\n\u003Ctd>August 2026\u003C\u002Ftd>\n\u003Ctd>December 2026\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>Deepfake labeling\u003C\u002Ftd>\n\u003Ctd>August 2026\u003C\u002Ftd>\n\u003Ctd>December 2026\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>Ban on intimate images without consent\u003C\u002Ftd>\n\u003Ctd>–\u003C\u002Ftd>\n\u003Ctd>December 2026\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>High-risk AI rules\u003C\u002Ftd>\n\u003Ctd>Postponed\u003C\u002Ftd>\n\u003Ctd>Postponed\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Fdiv>\n\u003Ch2>Who benefits – and who doesn't\u003C\u002Fh2>\n\u003Cp>Critics worry that primarily \u003Cstrong>tech giants\u003C\u002Fstrong> benefit from the delay. They gain time to further water down the planned regulations – industry has criticized the EU AI Act from the start as bureaucratic and innovation-hostile. Smaller providers and European companies that prepared for compliance could be disadvantaged.\u003C\u002Fp>\n\u003Ch2>What this means for German companies\u003C\u002Fh2>\n\u003Cp>The delay is a signal: the EU AI Act won't be enforced as planned. German decision-makers shouldn't wait for high-risk rules to arrive – they should prepare now. Because postponed rules eventually come. Early movers gain competitive advantage over those betting on the next delay. At the same time, it's worth watching lobbying activities: which rules will face pressure next?\u003C\u002Fp>\n\u003Ch2>Sources\u003C\u002Fh2>\n\u003Cul>\n\u003Cli>\u003Ca href=\"https:\u002F\u002Fwww.deutschlandfunk.de\u002Feu-ai-act-ki-verordnung-regeln-100.html\">Deutschlandfunk\u003C\u002Fa>\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Cp>\u003Cem>Editorially owned by \u003Ca href=\"\u002Fen\u002Fautor\u002Fideal-syka\">Ideal Syka\u003C\u002Fa>. Sources and method: \u003Ca href=\"\u002Fen\u002Fredaktion\">Newsroom & method\u003C\u002Fa>. Tips and corrections: \u003Ca href=\"mailto:ai@i6eal.de\">ai@i6eal.de\u003C\u002Fa>.\u003C\u002Fem>\u003C\u002Fp>\n",1785666620786]