The European Union has begun enforcing its groundbreaking artificial intelligence legislation, with the first batch of restrictions under the AI Act coming into effect on Sunday, despite mounting pressure from the US not to push ahead.

The law, which formally entered into force in August 2024, introduces prohibitions on AI systems deemed to pose "unacceptable risk" to citizens, including social scoring systems, real-time facial recognition, and AI tools that categorise people by race, sexual orientation, and other attributes.

Companies found in breach of the regulations face substantial penalties of up to €35 million or 7 per cent of their global annual revenues, whichever is higher - exceeding the fines possible under Europe's GDPR privacy law.

Tasos Stampelos, head of EU public policy and government relations at Mozilla, described the AI Act as "very much needed" despite not being "perfect." He emphasised that compliance will depend on forthcoming guidelines and secondary legislation that will "actually stipulate what compliance looks like."

The enforcement launch comes amid growing tension with the United States, particularly following Donald Trump's recent warnings about retribution for the EU's targeting of US tech companies. Trump has explicitly stated that he views moves by Brussels against US companies as "a form of taxation," expressing "very big complaints with the EU" during remarks at Davos.

However, a senior EU official involved in the AI Act's implementation told the Financial Times that while they acknowledged the US pressure, the law would not be altered. "What we can do is ensure that it is as innovation-friendly as possible, and that's what we're doing right now," the official said.

Steve Lester, chief technology officer at consultancy firm Paragon, emphasised the impact on businesses: "For all businesses that operate on the EU market, transparency and ethical AI practices will be absolutely key. In customer communications, this means going beyond simply disclosing AI use. Companies need to ensure their targeting and personalisation strategies completely avoid newly prohibited practices like manipulative AI systems, social scoring, or biometric categorisation."

The implementation will continue to roll out gradually, with further provisions targeting large AI models and high-risk AI-powered products being introduced through 2027. The European Commission is expected to publish crucial guidance on how these rules should be applied by companies on Tuesday.

---