Europe’s artificial intelligence law has reached its main application date, but its toughest rules have been pushed back. The EU AI Act, Regulation 2024/1689, entered into force in August 2024 and sorts artificial intelligence systems by risk, from unacceptable practices that are banned outright to minimal-risk tools that face almost no duties. Its general application date arrived on August 2, 2026, bringing transparency duties into force: people must be told when they are dealing with an AI system, and synthetic content must be labelled as artificially generated. The Digital Omnibus amendment, Regulation 2026/1744, entered into force on July 27, 2026, and deferred the demanding obligations for high-risk systems, moving Annex III uses to December 2, 2027, and AI built into regulated products to August 2, 2028. The same amendment added a ban on AI tools that create non-consensual intimate images, applying from December 2, 2026. Rules for general-purpose AI models have applied since August 2, 2025.
What the Evidence Shows
High-risk systems are the heart of the law, and they cover uses that shape people’s lives: screening job applicants, deciding access to credit, running exams, policing and delivering essential services. Providers and users of such systems will have to manage risks, keep records, ensure human oversight and monitor performance, and register the system in a European database. Critics argue the burden will slow European innovation and push developers to launch elsewhere, while supporters say the delay already handed industry more time and the rules are the price of public trust. The United States has no comparable federal AI law, so companies selling into Europe must meet the European standard anyway. For businesses, the work starts now: take stock of every AI system in use, train staff in what the law calls AI literacy, and label AI-generated content. The postponed deadlines are not a reprieve, they are a timetable, and December 2027 will arrive faster than it looks. Reporting and analysis from sources such as Reuters track the same developments, and specialists following EU AI Act implementation say the direction is now consistent across independent lines of evidence.
The Wider Context
The wider context is a public conversation conducted at a speed institutions were not built for. Developments around EU AI Act implementation are announced, contested and revised within a single news cycle, and the officials involved have learned to publish evidence alongside claims so that scrutiny can actually occur. Trust, the polling suggests, follows that transparency more than it follows confidence.
What Happens Next
What happens next in the story of EU AI Act implementation will be decided in the places where it is being implemented, measured and argued about in public. The officials and organisations involved have committed to publishing progress data, and that commitment gives residents, businesses and journalists something firmer than promises to judge. The next milestones are dated, which means they can be checked. You can follow related developments in our coverage of news.
There is a practical lesson in all of this for readers deciding what to do today. Start with the verified facts, be wary of anyone promising certainty, and prefer reversible steps while the evidence is still developing. Circumstances differ by country, income and personal situation, so general guidance should always be weighed against your own context before you act on it in any significant or lasting way.