By Artificial intelligence is rapidly escalating the risk of both large-scale cyberattacks on financial infrastructure and highly targeted fraud capable of draining individual bank accounts—and experts warn that banks must move faster to keep up.
Why This Matters for Your Financial Security
Banks have long been considered the gold standard in cybersecurity. But in an era of AI-powered threats, that reputation is under serious pressure.
“AI is making fraud so convincing and so cheap to pull off that we need to rethink how money is stored from the ground up,” said Laura Spiekerman, cofounder of fraud-prevention firm Alloy, in an interview with Axios.
Fraudsters Are Already Ahead of the Banks
Multiple cybersecurity and fintech experts speaking at a financial infrastructure panel at the HumanX conference noted that bad actors are currently outpacing financial institutions in the AI arms race.
What Cryptocurrency Security Can Teach Traditional Banking
Spiekerman pointed to the cryptocurrency ecosystem as a practical security model that holders of fiat currency could adopt:
- Hot wallets — connected to software and used for everyday transactions
- Cold wallets — kept entirely offline for longer-term savings storage
The advantage of this approach goes beyond simply splitting risk across multiple wallets. Each wallet can be assigned its own authentication protocols and security layers, providing layered protection that is harder to compromise in a single attack.
Practical Fixes Banks and Consumers Can Implement Now
Juan Pablo Ortega, co-founder of global payments platform Yuno, outlined several actionable steps that financial institutions can take immediately:
- Mandate passkeys rather than making them an opt-in feature
- Add friction to large transactions, such as secondary verification steps, to create an additional layer of defense against unauthorized transfers
The Bigger Threat: AI-Powered Attacks on Financial Infrastructure
While everyday AI fraud is a growing concern as models become cheaper and more capable, experts emphasize that the more significant danger lies at the infrastructure level—attacks targeting the backbone of the global financial system.
This is not a theoretical risk. Policymakers and major financial institutions are already war-gaming worst-case scenarios involving AI-enabled attacks on critical banking systems.
Anthropic’s Mythos Model Triggers Emergency Wall Street Meeting
In a significant development, Anthropic has begun a tightly controlled release of its Mythos model—described as the first AI system capable of crippling parts of the internet, including major financial institutions.
The release prompted an urgent high-level meeting: Treasury Secretary Scott Bessent and Federal Reserve Chair Jerome Powell convened with top Wall Street executives to assess the risks. The meeting, first reported by Bloomberg and confirmed by Axios, centered on the threat of AI-powered cyberattacks on bank systems and the preventative measures needed to counter them.
The Bottom Line
AI is reshaping the threat landscape for personal and institutional finance alike. From advanced phishing scams to possible attacks on critical systems, the financial sector is dealing with new types of risks that require quicker responses, better security measures, and a complete reevaluation of how money is kept safe.