EU May Pause Parts Of AI Law After Pressure From US And Big Tech

The European Commission is looking at delaying parts of its Artificial Intelligence Act after facing strong pressure from US technology companies and the American government. According to the Financial Times, the proposal forms part of a wider simplification package expected to be announced on 19 November.

The AI Act, which came into force in August last year, is seen as the world’s strictest set of rules governing artificial intelligence. It was designed to regulate how companies use and develop AI, especially in high-risk areas that affect health, safety or citizens’ rights. However, most of its key rules will only take effect in August 2026.

Officials in Brussels have been holding talks with Washington as part of this process. A senior EU official told the Financial Times that the bloc had been “engaging” with the Trump administration on changes to the act and other digital laws. The talks reflect fears in Europe that stricter rules could lead to US retaliation, including limits on intelligence or defence cooperation.

What Changes Are Being Considered?

A draft seen by the Financial Times shows that Brussels may give companies using high-risk AI a 1 year “grace period” before penalties apply. This would allow them more time to adapt without disturbing the market. The proposal is still being debated within the Commission and among European governments, so details could change before the final decision.

The delay could also affect when fines are enforced for breaking transparency rules. Under the new idea, these penalties might only come into effect in August 2027, giving AI developers and users extra time to meet the requirements.

The draft is also meant to simplify the process for companies by reducing admin work and centralising enforcement through a new AI office. Those in support believe this could make it easier to follow the law while keeping oversight within EU control.

How Are Companies And Lawmakers Reacting?

Tech companies such as Meta have argued that the AI Act might block Europe from accessing advanced tools and services. They claim that strict regulation could make the region less competitive against the US and China.

An EU spokesperson confirmed that talks were continuing over potential delays to “targeted parts” of the act but said the Commission still backed its main goals. The pause would not mean the law is being abandoned, but rather that enforcement would be adjusted to give companies more time.

If approved, the proposal will need approval from a majority of EU countries and the European Parliament. The final outcome is expected to show how they plan to balance control of AI with pressure from global powers.

Ben Gibson, UK CEO of Cosmo5 commented on the matter, saying: “The EU’s move to reconsider the timing of its AI Act is a recognition that innovation doesn’t advance under premature constraints. This is a technology still finding its shape, and creating the space to experiment is not a step back, it’s a step toward long-term strength.

“While AI is becoming more deeply embedded in daily life, we’re still very much in the phase where genuine value is discovered through testing, not by adhering to strict frameworks. Loosening the timeline gives smaller players and startups the room to explore how AI can truly change behaviour, not just comply with frameworks built for yesterday’s risks.

“The UK’s new AI lab reflects that same principle: progress depends on safe, open experimentation. If Europe wants to prove it can lead, not just regulate, this is the path that signals genuine ambition.”