July 6, 2026, (Inside AI) — A top UK financial regulator has called for a formal review of whether large language models like ChatGPT, Claude, and Gemini should be brought under regulatory oversight, as these general-purpose AI tools increasingly shape consumer financial choices.
Financial Conduct Authority (FCA) Executive Director Sheldon Mills made the recommendation on Monday, urging the watchdog to examine the “scale, nature and impact” of AI models that currently fall outside its remit. The proposal comes amid a broader FCA review into AI’s footprint in finance, which uncovered a striking reliance on unregulated AI for money guidance.
The review found that more than a quarter of UK consumers now trust tools from OpenAI, Anthropic, and Google for financial advice, even though few realize that regulated protections do not apply to these services. Mills recommended the FCA consider whether to “secure and adapt” its regulatory perimeter within the next three to six months.
“We need to keep pace with a rapidly changing environment and the principles‑based, outcomes focussed approach we’ve taken on AI,” said Ashley Alder, Chair of the FCA.
OpenAI, Anthropic, and Google did not immediately respond to requests for comment.
The push reflects a global scramble by regulators to address AI risks, from cyber threats tied to frontier models like Anthropic’s Mythos to the rise of agentic systems that operate with minimal human intervention. Mills’ review highlights how quickly consumer behavior is outpacing rulebooks written for a pre‑generative‑AI era.
The Quiet Drift into Uncharted Territory
Large language models are not designed as financial advisors, yet millions are using them as such. The FCA’s finding that trust outstrips awareness of missing safeguards signals a dangerous gap. Unlike regulated advisors or robo‑advisors, these chatbots offer no mandated transparency, no suitability checks, and no recourse when advice goes wrong.
The review warns that this gray zone could undermine the UK’s carefully built consumer protection framework. If a chatbot suggests a high‑risk investment to a vulnerable user, there is no clear liability. Mills’ call to review the perimeter is a direct attempt to close that gap before it widens.
This is not the first time technology has leapfrogged regulation. The 2008 financial crisis showed how unregulated shadow banking could destabilize markets. Today’s AI tools, while not yet systemic, are being embedded into household financial decisions at an unprecedented speed. The FCA’s move signals a desire to avoid playing catch‑up.
When Shared Code Becomes Shared Risk
The FCA review also flags a less visible danger: concentration risk. A recent survey found that 81% of global financial firms are adopting AI, with 40% at advanced stages of scaling. While most use remains in back‑office functions, British firms are increasingly deploying AI in customer‑facing roles, including complaint handling and investment guidance.
Mills warned that widespread reliance on a handful of tech providers could create correlated behavior and common points of failure. If multiple firms use the same underlying model or cloud infrastructure, a single glitch or adversarial attack could ripple across the financial system.
This echoes concerns raised by the Bank of England and international bodies like the Financial Stability Board. The review stops short of proposing specific limits but makes clear that the existing rulebook must evolve. The FCA’s principles‑based approach, which focuses on outcomes rather than prescriptive rules, will be tested as AI becomes more autonomous.
For now, the spotlight is on the three‑ to six‑month window Mills has proposed. If the FCA moves to regulate general‑purpose models, it could set a precedent for other jurisdictions. The UK has long prided itself on balancing innovation with protection. This latest review suggests that balance is about to be recalibrated.
