The European Union has firmly rejected calls from major tech companies, including Alphabet and Meta, delaying the implementation of the Artificial Intelligence Act. Many large tech companies, as well as several European AI companies, such as ASML and Mistral, have urged the committee to retain implementation, citing high compliance costs and complex regulatory requirements. However, the EU has now made clear that the legal timeline for deploying AI laws remains unchanged.
“Indeed, we've seen a lot of reports, lots of letters and lots of things being said about AI law. Let's be as clear as possible. There is no clock. There is no bounty period. There is no suspension. The committee also highlighted the legally binding deadlines that future laws will be adopted and implemented accordingly to ensure responsible AI development.
Why tech companies want delays due to AI law
Tech and AI companies are resisting the implementation of AI regulations by the EU in response to concerns that the law could curb innovation and increase costs. US tech giants, including Google's parent alphabet and Facebook owner Meta, have joined European companies urging Brussels to consider delays.
However, the committee is determined and even warns that delaying implementation will undermine public trust and security. “We have legal deadlines established in the legal text,” Regnier pointed out, outlining the provisions of the law and outlined that it will take effect in stages starting this year.
The EU also plans to simplify some digital rules later in 2025. This could reduce reporting obligations for small businesses. However, the committee made it clear that these simplifications do not affect the timeline of deployment or enforcement of the AI Act itself. With this law, the committee wants to place guardrails around AI. This is a rapidly advancing technology that currently touches almost every sector of the economy.
What is the EU AI Law?
The EU Artificial Intelligence Act is described as the world's first comprehensive regulatory framework for AI. It was officially enacted on August 1, 2024. The legislation includes guidelines for AI companies to follow a risk-based approach, categorizing AI systems based on pause level. The most dangerous uses are manipulative behavioral prediction tools and scraping of face recognition that is not targeted, completely prohibited under this law.
When will the AI Act come into effect?
The first set of regulations came into effect on February 2, 2025 and covers the most harmful AI practices. By August 2, 2025, the general AI (GPAI) model obligations will come into effect. For models that were already on the market by this date, providers should ensure compliance until August 2027. Meanwhile, rules governing high-risk AI systems, such as those used in employment, education, healthcare, or critical infrastructure, will come into effect on August 2, 2026.
– end
