Bank of England Governor Warns G20 That Advanced AI Poses Risk to Global Financial Stability
Andrew Bailey, writing in his capacity as chair of the Financial Stability Board, told international finance ministers that frontier AI systems are developing capabilities that could spread financial disruption across borders.
Key points
- Bank of England Governor Andrew Bailey sent a two-page letter to G20 finance ministers and central bank governors warning about AI risks to global financial stability.
- Bailey chairs the Financial Stability Board (FSB), the international body that monitors and makes recommendations about the world's financial system.
- The letter specifically flags "frontier" AI models, meaning the most advanced systems currently available, for showing sophisticated autonomous decision-making and, in Bailey's words, "threat capabilities".
- The warning focuses on the risk of cyber-disruption spreading across national borders, not staying contained within one country or institution.
What did Andrew Bailey actually say?
Bailey warned that the most advanced AI systems are no longer just useful tools. They are becoming capable of acting independently and solving complex problems in ways their designers did not specifically programme.
In a letter reported by The Guardian, Bailey described these "frontier" AI models as displaying "increasingly sophisticated autonomy and problem-solving abilities, as well as threat capabilities". That phrase matters. Bailey is not warning about AI making bad loans. He is warning that advanced AI could be used to attack financial systems, and that the damage would not stop at one country's border.
The letter went to finance ministers and central bank governors from G20 nations, the group of the world's twenty largest economies.
Why does a cross-border risk matter more than a local one?
Financial systems are deeply connected. A disruption in one major bank or payment network can ripple outward to others in minutes.
Bailey's role as chair of the Financial Stability Board, a body set up after the 2008 financial crisis to coordinate global financial oversight, gives the warning institutional weight. The FSB does not write law, but its recommendations shape national regulation across member countries.
The concern here is that a cyber-attack, meaning criminals or state-backed groups using computers to sabotage or steal, powered by advanced AI could move faster and adapt more cleverly than current defences expect. Worse, it could hit multiple jurisdictions, meaning multiple countries or regulatory zones, at the same time.
What should ordinary people take from this?
For now, this is a warning from a senior regulator to other senior regulators. Your bank account is not under immediate threat.
But warnings at this level tend to precede regulatory action. The FSB has previously issued guidance that shaped rules on bank capital requirements and oversight of financial technology firms. Expect formal FSB consultation papers or proposed guidance on AI risk management in financial services to follow, likely with comment periods open to industry.
If you work in financial services, insurance, or any regulated sector, watch for your own regulator citing FSB recommendations as the basis for new compliance requirements. That is how FSB guidance typically travels into national law.
Common questions
Is the government going to restrict AI because of this?
Not immediately. Bailey's letter is a warning to peers, not a legislative proposal. Regulatory rules in most countries require formal consultation, a proposed rulemaking stage, and an open comment period before any restrictions take effect.
Does this affect how banks use AI today?
Banks already use AI for tasks like fraud detection and credit scoring. Bailey's concern is specifically about the newest, most powerful AI systems and their potential to be used against financial infrastructure, not about routine AI already in use.
