The UK government’s AI Safety Summit was billed as a landmark moment for global regulation. But behind the scenes, a quieter, more insidious process is shaping the rules: a lobbying loophole that allows foreign tech giants to exert undue influence over British policy.

At the heart of this issue is the Advisory Board on AI Regulation (ABAIR), a quango created in 2022 to provide expert guidance to the Department for Science, Innovation and Technology. Its membership includes academics, civil society representatives, and industry figures. But a British Wire investigation has found that nearly half of its industry members are executives from US-based tech firms, including Google, Microsoft, and OpenAI. These individuals are not UK citizens, nor are their companies headquartered here. Yet they hold voting rights on policy recommendations that are often adopted wholesale by ministers.

The loophole is simple. Under current rules, there is no cap on the number of foreign nationals serving on such advisory bodies. A 2019 Cabinet Office review recommended that non-UK residents should not exceed 20% of members on public boards, but this has never been enforced for ABAIR. The result is that US tech interests effectively have a seat at the table when it comes to writing the UK’s AI rulebook.

Dr. Eleanor Shaw, a former advisor to the Centre for Data Ethics and Innovation, told British Wire: “The problem is structural. These companies have deep pockets and a strategic interest in ensuring that regulation remains light-touch. They frame their input as technical expertise, but it’s lobbying by another name. The UK is sleepwalking into a regime that prioritises corporate convenience over public safety.”

Documents obtained under Freedom of Information reveal that ABAIR’s draft recommendations on high-risk AI classification were significantly softened after interventions from US members. One internal email from a Google executive thanked the board for “accommodating” concerns about over-burdening innovation. The final report omitted key proposals for mandatory audits and liability clauses.

Transparency campaigners say the practice extends beyond AI. The UK’s Office for Investment, which fast-tracks projects from foreign firms, operates with minimal oversight. But AI regulation is particularly vulnerable because it is technical and fast-moving. Ministers rely heavily on industry input to stay abreast of developments, creating a dependency that savvy lobbyists exploit.

“The government has outsourced its thinking to the very companies it should be regulating,” said MP Clara Osborn, a member of the Commons Science and Technology Committee. “We have to ask: whose interests are being served? If the goal is to make Britain a global AI leader, that’s fine. But not at the expense of democratic accountability.”

The response from the Department for Science, Innovation and Technology was dismissive. A spokesperson said: “We are committed to building a robust AI regulatory framework that balances innovation with safety. Our advisory boards draw on a wide range of expertise, including from international partners. All members act in an individual capacity, not as representatives of their employers.”

But critics point out that these individuals are paid by their companies, attend meetings on company time, and often have their travel expenses covered by their employers. The line between personal expertise and corporate interest is blurred.

There is a growing movement to close the loophole. The Trade Justice Movement and the AI Now Institute are among the organisations calling for mandatory declarations of conflicts of interest and a cap on non-UK residents on advisory boards. The Scottish Parliament has already passed similar rules for its own expert panels.

Yet the government shows no sign of changing course. Instead, it has expanded ABAIR’s remit to cover enforcement of the upcoming AI Bill. That means the same foreign tech executives who helped write the rules will now oversee their implementation.

The question rarely asked is: why would a US company invest so heavily in influencing UK policy? One answer lies in the global race to set standards. If Britain adopts a permissive model, other countries may follow. The UK becomes a testbed for deregulation.

“The US tech sector sees London as a laboratory,” said Dr. Shaw. “They can trial softer rules here, then export that model to Brussels and beyond. The British public is being used as a guinea pig.”

Meanwhile, the meetings continue. ABAIR’s next session is scheduled for next month, behind closed doors. The minutes will not be published for two years.