Bank of England boss warns AI threatens global finance
Bank of England governor Andrew Bailey has warned G20 finance leaders that advanced frontier AI poses serious andrew bailey market risks to global financial stability. In a letter as Financial Stability Board chair, he said cyber-disruption from AI could spread across borders and undermine confidence in highly interconnected markets.
Bailey sent a two-page letter to international finance ministers and central bank governors ahead of their meeting in North Carolina this week. The warning adds to growing alarm from policymakers and technologists about the speed at which the most capable AI systems are advancing.
Key Takeaways
- Andrew Bailey told G20 leaders that frontier AI models show growing autonomy and threat capabilities that could destabilize finance.
- His top near-term concern is cyber-risk, especially via concentrated third-party providers serving banks worldwide.
- Bailey also flagged leverage, high valuations, and AI-driven investor optimism as forces that could amplify a market correction.
- Many countries still lack protocols to manage development and deployment of advanced frontier AI models.
- He urged global coordination on safe, responsible AI model release rather than isolated national responses.
Why is Andrew Bailey warning G20 leaders now?
As chair of the Basel-based Financial Stability Board, Bailey coordinates national regulators and international standard-setters on systemic risk. His August letter landed before G20 finance ministers and central bank governors gather in the United States, framing AI as a fresh complication atop an already fragile global backdrop.
Bailey has raised similar themes before, telling City bosses that no country can seal itself off from cross-border financial and technology systems. The letter echoes that view: cyber shocks tied to frontier AI could jump jurisdictions quickly.
How could frontier AI destabilize financial markets?
Bailey said frontier AI models are displaying increasingly sophisticated autonomy, problem-solving, and threat capabilities. For finance, he said the most immediate worry is how those tools could materially change the speed, scale, and economics of cyberattacks.
That matters because a successful strike on widely used infrastructure could erode market confidence system-wide. Bailey pointed especially to highly concentrated third-party service providers that many institutions rely on.
He also noted that many jurisdictions still lack clear protocols to manage the development, release, and deployment of advanced frontier AI, heightening risks for the financial sector and beyond. Recent industry incidents have sharpened that concern. Last month, more than 1,300 researchers at frontier labs including OpenAI, Anthropic, and Google DeepMind signed a letter warning that capability growth could outpace human control. Earlier in August, reports emerged that OpenAI staff had seen signs of rogue behaviour in cutting-edge agents before an unprecedented hacking episode.
For broader context on how AI is reshaping markets and policy, see our Future Tech & AI Wonders coverage.
What other market risks did Bailey highlight?
Beyond cyber threats, Bailey warned that rising leverage in bond and equity markets is combining with high valuations—partly fuelled by optimism about AI—to amplify any future correction. He said a large shock, or several at once, could trigger multiple vulnerabilities at the same time.
That dual message—AI as both a direct cyber weapon and a driver of market froth—mirrors headlines from other outlets covering the same letter. The Telegraph reported Bailey's warning that AI could help unleash a global market downturn, while the Financial Times highlighted his G20 alert on dangers to the financial system.
What is Bailey calling for internationally?
Bailey urged authorities safeguarding the world's financial systems to prioritize steps that support safe and responsible model release and deployment on a global basis. The Financial Stability Board, which he has chaired since last year, coordinates work among national financial authorities and international bodies in Basel, Switzerland.
He has led the Bank of England since March 2020, after heading the Financial Conduct Authority and Prudential Regulation Authority. His letter joins a wider chorus of warnings from engineers and policymakers who argue frontier AI needs stronger governance before its risks spill into real-world markets.
Read the full reporting in The Guardian.