EU waters down AI rules and delays key provisions until 2027

The European Union has agreed to significantly weaken its landmark artificial intelligence regulations, pushing back key provisions by over three years in what critics say shows Europe backing down to pressure from major technology companies.

EU countries and European Parliament lawmakers reached a tentative agreement Thursday after nine hours of negotiations to water down the AI Act, which originally entered into force in August 2024. The changes need formal approval from EU governments and the European Parliament in the coming months.

The most significant change delays rules for high-risk AI systems until December 2027. These systems include those involving biometrics, critical infrastructure, and law enforcement applications. The original deadline was August 2024, meaning companies now have more than three additional years to comply.

“Today’s agreement on the AI Act significantly supports our companies by reducing recurring administrative costs,” said Marilena Raouna, Cyprus’s deputy minister for European affairs. Cyprus currently holds the rotating EU Council presidency.

The changes reflect broader European Commission efforts to simplify digital regulations after businesses complained that overlapping rules and red tape were hampering their ability to compete with U.S. and Asian rivals. This complaint has become increasingly common as European tech companies struggle to match the scale and innovation speed of Silicon Valley giants and Chinese firms.

The agreement also excludes machinery from the AI Act entirely, following lobbying from major European industrial companies like Germany’s Siemens and Dutch semiconductor equipment maker ASML. These companies argued their products were already subject to existing sectoral regulations.

However, negotiators did agree on some new restrictions. The deal includes a ban on AI practices that create unauthorized sexually explicit images, responding to recent controversies involving deepfake content. This addresses issues with AI-generated intimate images, including those produced by systems like Elon Musk’s xAI chatbot Grok. The ban takes effect December 2024.

“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,” said Dutch lawmaker Kim van Sparrentak, who worked on the negotiations.

The agreement also requires mandatory watermarking of AI-generated content starting December 2024, helping users identify when they’re viewing artificial content.

The watered-down rules highlight the ongoing tension between regulation and innovation in the AI space. While the EU was first to create comprehensive AI legislation, the extended timeline raises questions about whether Europe can maintain its regulatory leadership as AI technology advances rapidly.

Industry reactions were mixed. The European Consumer Organisation criticized the weaker protections, while tech lobbying group CCIA said lawmakers should have gone further in reducing regulatory burden.

Even with these changes, EU officials maintain their AI rules remain the strictest in the world. The regulations were originally designed to address concerns about AI’s impact on children, workers, companies, and cybersecurity. The question now is whether the extended timeline will allow other regions to catch up or surpass Europe’s regulatory framework.

The development comes as global competition in AI intensifies, with the United States and China racing ahead in both AI development and deployment. European policymakers face the difficult balance of protecting citizens while ensuring their companies can compete in the global AI market.