New AI models pose growing threat to the global economy, Bailey warns
Bank of England Governor Sounds Alarm to G20 as Autonomous AI Systems Threaten Financial Infrastructure
Poinews.com – The intersection of Middle East conflict, sweeping trade disruption, and a new generation of self-directed artificial intelligence systems has pushed the head of the Bank of England to issue a formal warning to the world's most powerful economies. In a letter dated 28 August, Andrew Bailey — who simultaneously chairs the Financial Stability Board (FSB) and serves as Governor of the Bank of England — addressed G20 finance ministers and central bankers ahead of their two-day gathering in Asheville, North Carolina, United States, on Monday and Tuesday.
The timing of the intervention is deliberate. Global markets are already absorbing shocks from multiple directions at once. Ongoing conflict in the Middle East has pushed energy prices higher and stoked inflationary pressures across consumer economies. Simultaneously, the scale of capital being poured into AI development within the United States is adding its own inflationary weight to an already strained macroeconomic picture. Layered on top of these forces are unprecedented trade decisions by US President Donald Trump, which have rendered the international commercial environment markedly less predictable than at any point in recent decades.
A New Class of Threat: Autonomous Models and Digital Infrastructure
Into this already turbulent backdrop, Bailey argues, is arriving a qualitatively different kind of risk. A new cohort of sophisticated AI models is now capable of locating and exploiting weaknesses inside complex digital systems — banking cores, payment rails, and other sensitive server architectures — with consequences that could cascade through the broader economy. The concern is not merely that these systems might malfunction, but that they could be deployed, or deployed against, in ways that destabilise market confidence at scale.
The most cited illustration is Mythos, a model produced by Anthropic that combines advanced autonomous coding with cybersecurity capabilities. Unveiled in April 2026, Mythos has been withheld from public use precisely because of the magnitude of risk its capabilities represent. Its existence underscores a point Bailey makes explicitly: the frontier of AI development is moving faster than the regulatory and resilience frameworks that jurisdictions have built to contain technological disruption.
"Frontier AI may have the ability materially to alter the speed, scale and economics of cyber risk, which could undermine market confidence system-wide, especially due to highly concentrated third-party service providers."
That concentration is a critical detail. A handful of cloud and infrastructure providers underpin a disproportionate share of global financial operations. If an autonomous model were to identify and exploit a vulnerability in one of those shared platforms, the blast radius would extend far beyond any single institution or even any single country.
Borders Are Invisible to Algorithms
Perhaps the most consequential observation in Bailey's letter is the reminder that digital disruption does not pause at customs checkpoints. National governments retain the authority to legislate, supervise, and enforce within their own territories, yet an AI-driven incident with financial-stability implications can propagate globally from the moment it begins, regardless of where it originates.
"AI will not respect national borders."
The practical implication, as Bailey frames it, is that regulatory asymmetry itself becomes a vector of systemic risk. If one jurisdiction has robust cyber-recovery capacity and another does not, an incident in the weaker jurisdiction can transmit stress into the stronger one through shared infrastructure, correspondent banking relationships, and interlinked market positions.
"Differences in legal frameworks, cyber capability, resilience and recovery capacity across jurisdictions could therefore have consequences well beyond the jurisdiction in which an incident originates and may themselves become a source of vulnerability."
The FSB's Stance and the Call for Coordinated Action
Writing in his capacity as FSB chair, Bailey reaffirmed the body's operational mandate. The organisation will continue to scan for emerging vulnerabilities, work to harden the resilience of financial systems, and ensure that technological innovation proceeds in a manner compatible with stability. He stressed that many jurisdictions still lack adequate protocols for managing the specific risks posed by frontier models, and that this gap must be closed before the next wave of deployments arrives.
"Taking appropriate steps to support safe and responsible model release and deployment on a global basis should in my view be a priority and would benefit all sectors of the economy, including by supporting financial stability and economic growth."
The letter also directed firms and supervisory authorities to recalibrate their internal safety architectures. Bailey urged preparation for what he described as a shifting threat landscape:
"A threat environment characterised by a higher volume of vulnerabilities and a faster pace of patching."
In practical terms, that means banks, insurers, and payment processors will need to accelerate their own patching cycles, stress-test for autonomous-attack scenarios, and coordinate with regulators on shared early-warning indicators. The FSB's role is to facilitate that coordination across borders, given that no single national supervisor can see the full topology of global digital financial infrastructure.
What Comes Next
The Asheville meeting gives G20 finance ministers and central bankers a forum to respond to Bailey's letter in real time, potentially shaping joint statements on AI governance, shared-resilience standards, and the sequencing of model releases. The next scheduled G20 finance gathering will take place in Miami, Florida, on 14–15 December, offering a further checkpoint at which progress — or the absence of it — on these questions can be assessed.
For market participants, the message is straightforward: the risk calculus for digital infrastructure is changing. The tools that build and run modern financial systems are themselves becoming objects of autonomous optimisation, and the institutions tasked with safeguarding those systems are being asked to move at a pace that matches the models they oversee. Whether that pace can be achieved before the next frontier deployment reaches production scale remains the central open question of the coming months.
Related Reading
Frequently Asked Questions
What is New AI models pose growing threat?New AI models pose growing threat is the main topic of this guide. The article explains the context, practical details, and next steps readers should understand.
Why does New AI models pose growing threat matter?New AI models pose growing threat matters because readers are looking for a useful answer, not just a short summary. Good content should match search intent and help them decide what to do next.