The International Monetary Fund has warned that AI-powered cyberattacks could trigger systemic disruption across global financial markets. In a blog published on 7 May, the IMF stated that “in the hands of bad actors, the next generation of AI will undermine financial stability”. The warning follows the unveiling of Anthropic’s Claude Mythos model which autonomously discovered thousands of previously unknown vulnerabilities across every major operating system and web browser.
Anthropic announced Claude Mythos Preview in April 2026 but has restricted access to roughly 50 companies through Project Glasswing, an industry consortium including Amazon, Apple, Microsoft and JPMorganChase. In pre-release testing, Mythos identified thousands of previously unknown zero-day vulnerabilities across every major operating system and every major web browser. It found flaws that had survived decades of human security review and millions of automated tests. In one documented case, the model uncovered a 27-year-old vulnerability in OpenBSD.
Machine Speed Attacks Will Outpace Human Defenders
The IMF warned that AI was dramatically lowering the cost and time needed for hackers to identify and exploit vulnerabilities, increasing the risk of systemic financial shocks. The fund’s analysis suggests that extreme cyber incidents could now spread faster than institutions can respond. “Extreme cyber-incident losses could trigger funding strains, raise solvency concerns and disrupt broader markets,” the IMF said.
The fund warned that a single vulnerability discovered and exploited across dozens of institutions could be devastating for the global financial ecosystem. This is not theoretical. The financial system relies on shared digital infrastructure that’s highly interconnected, including software, cloud services and networks for payments and other data. Advanced AI models can dramatically reduce the time and cost needed to identify and exploit vulnerabilities, raising the likelihood of simultaneously discovering and targeting weaknesses.
Bank of England governor Andrew Bailey captured the scale of the threat more bluntly, frontier AI systems could “crack the whole cyber risk world open”.
Anthropic’s Model Was Not Built for Attacks
Claude Mythos Preview was not built as a security tool. Anthropic built it to push the boundaries of software engineering. Its Claude Mythos zero-day discovery capabilities are an emergent property of its coding and reasoning depth. That distinction matters because it signals these capabilities will appear in general-purpose models from other AI labs, not just purpose-built offensive tools.
Independent testing by offensive security firm XBOW found that compared to Anthropic’s Opus 4.6, Mythos Preview cut false negatives by 42 percent. With additional source code access, that reduction hit 55 percent. The model reproduced vulnerabilities and developed working exploits on the first attempt in over 83% of cases.
Anthropic’s own team estimates that similar capabilities will proliferate from other AI labs within six to eighteen months. OpenAI is reportedly developing a model with comparable abilities. The exclusivity window is narrow.
Nordic Financial Institutions Face Concentrated Risk
Nordic financial institutions are particularly exposed to AI-powered attacks. Danske Bank is set to reduce the number of staff working in its retail operations across Nordic “home markets” Denmark, Sweden, Norway and Finland by up to 430 by the end of the year as AI transforms back-office operations. Tryg launched an ambitious investment plan in January to digitise all core areas of its Nordic financial services operations. A key part of this plan will involve the establishment of a Nordic AI hub to serve all branches of its activities across the region.
The region’s high digital adoption creates both opportunity and vulnerability. Nordic countries rely heavily on shared infrastructure including BankID systems, common payment rails and cloud platforms operated by the same handful of providers. A single well-targeted vulnerability could affect multiple institutions simultaneously.
Financial Regulators Must Act Now
The IMF urged regulators to treat cybersecurity as a core financial stability issue rather than simply a technical challenge, warning that the pace of AI development means authorities must move quickly to strengthen defences before risks escalate further. The fund called for greater international cooperation, stronger information sharing and improved cyber resilience standards.
Regional banks, credit unions and smaller financial firms are operating in the same threat environment as JPMorgan with a fraction of the security resources. The access divide will determine which institutions survive the next wave of AI-powered attacks.
Despite the risks, the IMF stressed AI could still become one of the financial sector’s most powerful defensive tools. Banks are increasingly deploying AI systems to detect fraud, identify vulnerabilities and accelerate responses to attacks. “When attackers operate at machine speed, defenders must do the same,” the fund said.
References
- IMF warns AI cyberattacks could trigger global financial crisis
- IMF Warning: AI Cyber Threats Could Shake Finance
- IMF warns of financial risks as AI boosts cyberattacks
- Anthropic’s Claude Mythos and What it Means for Security
- Claude Mythos becomes the first AI model to clear all cyberattack simulations
- AI driving changes in Nordic financial services
This post is also available in:
Svenska
Related News
Shadow AI Now Factors in One in Five Data Breaches
One in five data breaches last year involved AI tools that companies did not know their own staff were using. That figure comes from IBM's…
June 2, 2026 
GCHQ Chief Warns AI Is Being Weaponised Just Below the War Threshold
Artificial intelligence is being turned into a weapon that operates just below the threshold of open war, the head of GCHQ warned on Wednesday, and…
May 29, 2026 G7 Nations Release AI SBOM Framework to Map Supply Chain Dependencies
CISA and its Group of Seven partners released guidance this week outlining minimum elements for artificial intelligence software bills of materials, a framework that could…
May 15, 2026 