Brexit, Five Years On: Impact on UK and EU AI Regulation

Key Takeaways

• Brexit has had sweeping effects on AI regulation in the U.K. and EU.
• Leaving the EU five years ago means the U.K. has not had to adopt or implement the EU’s AI Act.
• Despite Brexit giving the U.K. freedom to create its own legislation, companies still need to comply with EU AI frameworks if they want to operate in the bloc.

Five years after Brexit, the impact of one of the biggest and most divided decisions in the U.K. is still only just emerging.

While the U.K. is experiencing a shift in the political landscape and new challenges for trade, much of the technology industry has seen positives in the country’s new-found freedom over artificial intelligence regulation.

Following Brexit, Britain has pursued a flexible and pro‑innovation approach to AI, while the EU remains committed to the comprehensive, binding rules of its EU AI Act.

However, the difference in regulatory approaches has created a dual environment for businesses, adding extra responsibility and cost to navigate both frameworks.

Post-Brexit U.K. AI

In short, Britain’s departure from the EU has meant the U.K. is no longer required to adopt or implement the EU’s AI Act.

This has given the U.K. government room to design a framework that it believes is better suited to promoting Britain.

In contrast to the one‑size‑fits‑all approach of the EU, the U.K. has been free to evolve its policy gradually with a lighter touch.

However, this means the U.K.’s AI regulation remains a work in progress.

“The UK approach aims to balance safety whilst not stifling innovation,” Amanda Brock, OpenUK CEO, wrote in a company outline .

“The government continues to gather evidence to understand the risks and challenges better,” Brock said.

In January, U.K. Prime Minister Keir Starmer announced a U-turn to the previous government’s approach to AI.

Under new leadership, Britain is now tackling AI closer to the way President Donald Trump has proposed—putting innovation at the front of the list of AI priorities.

This is a significant change from the previous Conservative government, which aimed to position the U.K. as a world leader in AI safety regulation

The aim was to put safety at the front of every piece of legislation; however, this was met with pushback from tech leaders who felt it risked putting the U.K. far behind the competition of the U.S. and China.

EU’s AI Approach

The EU AI Act categorizes AI systems by risk—unacceptable, high, limited, and minimal—and imposes strict obligations such as mandatory conformity assessments and severe fines for non‑compliance.

This top‑down framework is designed to cover all sectors uniformly, ensuring stringent protection of fundamental rights and heavy compliance burdens on companies.

While undoubtedly making a mark on AI safety, the EU AI Act has turned away some key players in the AI race, resulting in potential investment and business losses for member countries.

In 2024, both Meta and Apple chose not to distribute new AI products in the EU due to legislative uncertainty in the bloc.

U.K. Still Impacted by EU AI Act

Despite Brexit giving the U.K. freedom to create its own legislation, companies will still need to comply with the provisions of the EU AI Act if they want to operate in the EU.

This creates a dual‑regulatory environment where firms must navigate two different sets of rules, potentially increasing costs and complicating market access.

Derren Nisbet, CEO of London-based AI testing company Virtuoso QA, told CCN that Britain’s departure from the EU AI Act is both “positive and challenging.”

“The major issue is that we will have to comply with the U.K. framework and the EU one, and that is a hindrance to everyone,” Nisbet said.

“Any framework across the EU is challenging to get right; there are a lot of interested parties and cultural clashes,” he added. “I believe the GDPR act is an absolute disaster, hasn’t achieved anything really, and is treated with disdain by the U.S. tech giants it was aimed at,” he added.

While the U.K. government’s approach aims to be adaptable, there is an ongoing debate about whether a light‑touch approach to AI will sufficiently address emerging risks, or if it will actually make it more difficult for Britain to expand outwards.