EU AI Act News & Analysis
Plain-language guides to EU AI Act compliance — so you can spend less time on legal research and more time running your business.
Latest Articles
The AI Omnibus Trilogue Collapsed. Your August Deadline Is Back — For Now.
After 12 hours of negotiations on April 28, EU countries and the European Parliament failed to reach a deal on the AI Omnibus. Talks resume next month. Until the Official Journal publishes the agreed delay, August 2, 2026 is your legal deadline — and the uncertainty is back.
Read article →
The AI Omnibus Is Adding a 'Nudifier' Prohibition and SME Relief. Today Is Decision Day.
As trilogue negotiations conclude on April 28, EU lawmakers are set to prohibit AI 'nudifier' applications outright — and extend regulatory relief to mid-sized companies with up to 750 employees. Here is what SMEs need to know about both measures before the final text is locked in.
Read article →
EU AI Act Trilogue Closes In On Fixed Delay — What December 2027 Actually Means for Your Compliance Plan
The EU AI Act trilogue negotiations are entering their final stretch. A political agreement is expected tomorrow, April 28, and both the Council and Parliament have converged on hard postponement dates: December 2, 2027 for standalone high-risk AI systems, August 2, 2028 for AI in regulated products. Here is what that means practically for SMEs — and why the safest course of action right now is unchanged.
Read article →
The EU AI Omnibus Deal Is Almost Done — and It Probably Saves You 16 Months
With a political trilogue scheduled for April 28, both the Council and Parliament have independently settled on the same new high-risk deadlines: December 2, 2027 for standalone Annex III AI systems and August 2, 2028 for AI embedded in regulated products. The original August 2, 2026 deadline is very likely to move. Here is what that means for your compliance planning.
Read article →
Your Contact Center's AI Is Probably Analyzing Employees Illegally. And Customers Are Next.
The voice analytics platform your contact center runs might be two different legal objects at the same time: one that is already prohibited under Article 5 of the EU AI Act, and another that becomes a high-risk system in exactly 100 days. Here is the split-personality compliance problem that is landing on thousands of European businesses — and what you need to do before August 2.
Read article →
Ten EU Countries Just Pushed Back on Germany's Bid to Strip Industrial AI from the AI Act
Austria, Denmark, the Netherlands, Slovakia, Slovenia, Spain, Greece, Portugal, Romania, and Latvia have formally opposed a Parliament-led proposal that would move industrial AI products — machinery, medical devices, toys — out of the AI Act's horizontal framework and into sector-specific law. Germany is fighting hard for the shift. The final trilogue negotiation is next Tuesday. Here is what is at stake for EU SMEs.
Read article →
The EU AI Act's Hardest Line Is Already Live — And Most Businesses Haven't Noticed
While everyone focuses on the August 2026 deadline for high-risk AI systems, Article 5's eight prohibited AI practices have been enforceable since February 2025. Up to €35 million in fines apply now. Here's what is already illegal under EU law, and why the enforcement gap is wider than most businesses realise.
Read article →
In Six Months, the EU AI Office Gets Real Teeth. Here Is What That Means for Every Business Using GPAI Models.
The EU AI Act gave the AI Office enforcement powers over General-Purpose AI models — but those powers do not switch on until August 2, 2026. With six months to go, here is what the AI Office can actually do from that date, which providers are in the firing line, and what you need to have ready.
Read article →
Siemens and Germany’s Chancellor Are Demanding a Carve-Out for Industrial AI. What That Means for Every SME Using AI in Europe.
At Hannover Messe on April 19–20, 2026, Siemens CEO Roland Busch and German Chancellor Friedrich Merz publicly called for industrial AI to be exempt from the EU AI Act’s high-risk rules — threatening to redirect Siemens’s €1 billion AI investment to the US and China. It’s the most high-profile business challenge to the AI Act yet, and it’s landing four months before the August 2 enforcement date.
Read article →
The EU AI Office Cannot Access Anthropic's Most Powerful Model. That Is a Problem for Enforcement.
Anthropic's Mythos model, described as potentially the most powerful AI system yet announced, is off-limits to the EU AI Office. Eight AI safety groups are now demanding the EU grow its safety unit from 36 to 160 staff by 2030. Here is why this matters for the enforcement capacity the AI Act was supposed to have.
Read article →
GDPR Fines Hit €1.15 Billion in 2025. The AI Act Is Adding a Whole New Layer on Top.
The EDPB’s 2025 Annual Report, published April 9, 2026, documents a year of record GDPR enforcement — €1.15 billion in fines, a first-ever joint DMA-GDPR guideline, and a suite of practical compliance tools emerging from the Helsinki Statement. For EU SMEs already navigating AI Act obligations, the message is clear: GDPR enforcement is not slowing down — it’s coordinating with the AI Act and getting more structured about it.
Read article →
EU AI Act Regulatory Sandboxes: Most Member States Are Not Ready — But Yours Can Still Use Them
The August 2, 2026 deadline for AI regulatory sandboxes is approaching, but most EU Member States have not yet established theirs. An April 2026 European Parliament analysis reveals only one fully operational sandbox across all 27 Member States as of August 2025. Here is what that means for SMEs developing or deploying high-risk AI systems.
Read article →
Your AI Agents Are Generating Decisions Right Now. Can You Prove It? Article 12 and the Logging Obligation SMEs Can't Afford to Ignore
A fresh technical analysis published April 16, 2026 breaks down exactly what the EU AI Act requires from AI agent logging — and the answer is more specific and more demanding than most businesses expect. Automatic recording, tamper-evidence, six-month retention, and no finalised technical standard yet. Here is what every SME running AI agents needs to understand before August.
Read article →
GDPR's Transparency Net Is Already Closing on AI — 25 European Regulators Are Watching
While the EU AI Act high-risk deadline debate dominates compliance headlines, something quieter and more immediate is happening: GDPR enforcement on AI transparency is already live. In March 2026, the European Data Protection Board launched CEF 2026 — a coordinated enforcement action in which 25 national regulators will examine how companies disclose AI processing to individuals. Findings land in H2 2026.
Read article →
Your AI Agents Are Not Just Tools — The EU AI Act Treats Them Differently
While most EU AI Act coverage focuses on chatbots and hiring tools, the regulation has specific, harder rules for autonomous AI agents — systems that take actions without human input at every step. Here is what that means for the small and medium businesses that are quietly deploying them.
Read article →
The Real Cost of EU AI Act Compliance for SMEs: What the Numbers Actually Say
New compliance cost data published in April 2026 shows SMEs facing compliance ranges of EUR 50,000 to EUR 500,000 depending on complexity — and disproportionately higher per-employee burdens than large enterprises. Here is what the numbers mean in practice for small businesses using AI.
Read article →
Mistral's Wake-Up Call to Europe: 22 Measures to Avoid Becoming a "Vassal State" in AI
On April 7, 2026, Europe's highest-profile AI company published an extraordinary policy plea — 22 recommendations for how the EU should protect its technological sovereignty before it becomes dependent on American or Chinese AI systems for everything from commerce to defence. The message from Mistral's CEO Arthur Mensch was blunt: act now, or accept foreign domination.
Read article →
The EU Just Asked: Is ChatGPT a "Large Online Platform"? The Answer Could Affect Millions of European Businesses.
On April 11, 2026, the European Commission confirmed it is assessing whether ChatGPT has crossed the 45 million DSA user threshold for "large online platform" designation. If confirmed, OpenAI faces very significant new obligations — and the downstream effects on European SME users of ChatGPT are worth understanding.
Read article →
Civil Society Is Sounding the Alarm on the AI Act's Annex I Changes — With 17 Days to the Trilogue Deadline
As trilogue negotiations enter their final phase with a political agreement expected on April 28, CDT Europe and 32 civil society organisations have written to co-legislators warning that changes to Annex I of the AI Act could fundamentally reshape how AI is regulated across Europe's product sectors.
Read article →
The EU's AI Rule Has a Threshold Problem — And It's About to Catch Up With Foundation Model Providers
The EU AI Act uses a 2023 compute threshold to flag AI models as "systemic risk." As of April 2026, dozens of models are expected to blow past it — yet the Commission has not committed to updating the number. Here's why that matters for every business that uses or builds on top of foundation models.
Read article →
Your Employees Don't Need a Law Degree — But They Do Need AI Literacy Training. The EU AI Act Agrees.
As the Digital Omnibus simplifies other parts of the AI Act, lawmakers have drawn a firm line on one obligation: AI literacy training for employees who work with AI systems. Here's why it survived the simplification wave — and what it means for your team.
Read article →
The 'Wait and See' Window Is Closing — For a Reason Most Businesses Haven't Noticed
The AI Omnibus hasn't been finalised yet. But a quiet shift in how the law works means the 'wait and see' strategy was already losing ground before trilogue concluded. Here's what changed and why it matters for SMEs.
Read article →
The EU AI Act's 'Wait and See' Window Is Closing — Fixed Deadlines Are Here
While the Digital Omnibus delays high-risk AI obligations to December 2027, the more significant shift for compliance teams is subtler: both Parliament and Council have settled on fixed calendar dates instead of standards-linked triggers. That removes the last defensible reason to wait.
Read article →
The AI Act Has a Loophole — And the Delay Makes It Bigger
The EU AI Act is non-retroactive. Combined with the Omnibus delay pushing high-risk deadlines to December 2027, critics warn that AI systems deployed before then may never need to comply. Here's what that means for SMEs.
Read article →
EU AI Regulatory Sandboxes Are Free for SMEs — And Most Businesses Have Never Heard of Them
Every EU member state must have an AI regulatory sandbox running by August 2, 2026. They are free for SMEs and startups. A new European Parliament report published this week reveals the state of play — and why most countries are still scrambling to set them up.
Read article →
The AI Omnibus Enters Its Final Four Weeks — And a Deal by April 28 Could Change Everything
Trilogue negotiations on the AI Omnibus started the same day Parliament voted on March 26. The target date for a political agreement is April 28 — just four weeks away. Here's what that timeline means for businesses waiting on certainty.
Read article →
The EU's AI Law Kicks In August 2026. Most Countries Haven't Even Hired the Enforcers Yet.
The European Parliament just voted to delay high-risk AI compliance deadlines. But the stranger story is why: only 8 of 27 EU member states have set up the enforcement offices that were legally required since last summer.
Read article →
Trilogue Starts Next Month. Here's What the Three EU Institutions Are Actually Fighting Over.
With Parliament's vote locked in on 26 March, all three EU institutions now have their positions on the AI Act. Negotiations begin in April. Here's what they agree on, where they differ — and what SMEs should watch.
Read article →
The EU's Draft Rulebook for AI Content Labels Is Out. Here's What It Actually Requires.
The European Commission published a second draft Code of Practice on AI content labelling on 5 March 2026 — a practical guide to complying with Article 50's transparency rules. It's more flexible than the first draft. But the clock is ticking.
Read article →
Everyone Celebrated the AI Act Delay. Nobody Mentioned the November Deadline.
The EU Parliament voted to push high-risk AI rules to 2027. But the same vote set a firm November 2, 2026 deadline for labelling AI-generated content — the rule that will actually affect most small businesses.
Read article →
Experts Warn: EU Parliament's AI Act Delay Is Not a Reprieve
The EU Parliament voted this week to push high-risk AI deadlines to 2027–2028. But analysts from Gartner, FormerGov, and Info-Tech Research Group are unanimous: keep your compliance work going. Here's why — and what it means for SMEs.
Read article →
What EU AI Act Article 50 Means for Your Business: Chatbot Disclosures and AI Content Labels
While everyone's focused on the high-risk AI deadline getting pushed to 2027, there's a quieter obligation arriving on August 2, 2026 that affects far more businesses: Article 50's chatbot and AI content disclosure rules.
Read article →
EU Parliament Confirms AI Act Delay — High-Risk Deadline Pushed to December 2027
The European Parliament's plenary voted yesterday to officially adopt its position on the Digital Omnibus on AI. High-risk AI compliance is now set for December 2027. Here's what it means for SMEs.
Read article →
EU Parliament Votes Today on AI Act Simplification — Including a New Ban on Deepfake Nudifier Apps
Today's European Parliament plenary vote finalises MEPs' position on simplifying the AI Act. High-risk deadlines pushed to 2027-2028, nudifier AI banned, SME support extended. Here's what it means.
Read article →
High-Risk AI Deadlines Extended — What the EU Actually Agreed
Parliament and Council have agreed new fixed deadlines: December 2027 and August 2028 for high-risk AI. Here's what changed, what didn't, and what it means on Monday morning.
Read article →
The EU Just Made AI Compliance Easier for SMEs
On March 13, the EU Council voted to simplify the AI Act. Lighter documentation, possible deadline extensions, softer AI literacy rules. Here's what actually changed — and what didn't.
Read article →
The August 2026 Deadline Most Businesses Are Getting Wrong
Most businesses are planning for 2027. The real deadline is August 2026 — and it's already closer than you think.
Read article →
High-Risk AI: The Distinction Most Businesses Get Wrong
You bought an AI hiring tool. You didn't build it. But you're still on the hook. Here's how provider vs. deployer actually works — and which category your business falls into.
Read article →
5 Things SMEs Must Do Before the AI Act Deadline
A pragmatic to-do list from someone who's been through the compliance process. Urgent, specific, and no generic checklist fluff.
Read article →
The €35M Fine Is Real — But That's Not Your Real Risk
The headline fine number is real but misleading. Here's an honest look at what enforcement actually means for SMEs — and why reputational risk is probably a bigger deal than fines.
Read article →
The Chatbot Fix Is One Sentence. Here's the Harder Stuff.
Most businesses can make their chatbot compliant in five minutes. Then there are the edge cases — emotion detection, deepfakes, AI-generated content — that are actually more complicated.
Read article →
Know your EU AI Act risk level in 10 minutes
Our free audit walks you through the exact questions to classify your AI systems and identify what you need to do before August 2, 2026.
Start Free Audit →