Securing digital financial assets from AI-driven fraud, a shared mission

Artificial Intelligence (AI)-assisted fraud intersects three connected participants: consumers who need to protect digital access to their financial assets and virtual lives, banks with an obligation to authenticate customers and safeguard accounts, and federal banking agencies that promote a safe, sound and resilient financial system. Digital estate planning offers a useful example of this intersection by connecting bank controls, supervisory expectations and household preparedness into one risk management conversation.

For consumers, the risk is immediate and personal: AI can make impersonation faster, cheaper and more convincing, allowing cybercriminals to defraud individuals when they are settling an estate. For banks, the same fraud attempt becomes an operational and compliance challenge involving identity verification, elder exploitation and payment controls. As the nation’s central bank, the Federal Reserve supports financial stability, supervises and regulates banking organizations, promotes a safe and efficient payment system, and helps ensure banks treat consumers fairly. AI-assisted fraud can undermine confidence in banking, test banks’ operational resilience, increase losses to households and institutions, and if left unchecked at scale, contribute to broader financial system vulnerabilities.

AI-enabled elder exploitation: a growing threat

The scale of the threat is growing quickly. According to the FBI’s Crime Report, overall cybercrime losses exceeded $20 billion, a 26 percent increase from 2024, based on more than one million complaints. For the first time, the report included an AI-related descriptor: Americans filed 22,364 AI-related complaints and lost $893 million to AI-driven scams in 2025. This figure is likely conservative because it captures only cases in which victims recognized and reported the AI component.

Elderly individuals, who may be less familiar with AI-related scams yet often hold significant financial assets which they can access digitally, are particularly vulnerable. The FBI issued a Public Service Announcement warning of scammers targeting senior citizens after receiving complaints regarding grandparent scams resulting in at least $1.9 million in victim losses.

Unfortunately, scammers have become more sophisticated. They are using AI to mimic the voice or likeness of deceased or elderly individuals, aiming to intercept inherited financial accounts. These cases are personal for families, but they also test banks’ ability to verify identity, recognize exploitation and stop suspicious transfers before losses spread.

Deepfakes, agentic AI and the new fraud burden on banks

Several financial industry leaders described the challenges posed by AI-assisted fraud in a recent American Banker article. They include lower barriers to entry for committing fraud without expertise in financial or cybercrime and increasingly realistic personalization that makes scams easier to make and harder to detect. This allows economies of scale to be achieved within hours.

Evolving bank defenses: from real-time detection to consumer education

Banks are deploying AI-powered tools in new ways due to Section 314(b) of the USA PATRIOT Act which enables banks to share customer and transaction data with other banks free from civil liability. This collaboration connects fragmented institution-specific data in order to pool anomalous behavioral indicators, such as deep-fake patterns or automated bulk account takeovers, to rapidly update their AI-detection models. The 314(b) program enables banks to freeze accounts or stop fraudulent transactions in real time. This addresses a concern voiced by smaller community banks about a lack of timely response from their larger partners in responding to urgent information requests to facilitate processing or halt transactions.

Banks, industry groups and law enforcement agencies like the ABA and FBI, respectively, are also collaborating to develop educational resources to help customers identify deepfakes, verify authentic communications and protect personal data against advanced social engineering schemes.

Supervisory expectations for AI governance and risk management

U.S. Federal Banking Agencies including the Federal Reserve, OCC and the FDIC apply broad, principles-based risk frameworks to supervise evolving generative and agentic AI models. For example, the Financial Stability Board (FSB) published a report outlining sound practices for financial institutions in navigating benefits and risks responsibly as they adopt AI. On Sept. 2, 2026, the Federal Reserve Board released some updated resources that may be helpful to institutions as they manage the opportunities and risks of frontier AI for cybersecurity. The message for bankers is that innovation is expected to be paired with governance, controls, testing and accountability.

Consumer vigilance: protecting digital assets before crisis strikes

Consumers also have a role to play. In addition to digital keys and passwords, more financial apps have introduced Multifactor Authentication (MFA) to control access to customer accounts. MFA is a layered approach to securing data and applications where a user must present a combination of two or more credentials, such as a password and then a passcode sent to their phone, to verify their identity for login. It is essential that families secure MFA credentials in addition to passwords and other credentials for the financial apps that manage their savings and investments as part of their digital estate planning.

The offline key solution

Customers should designate a digital executor—a trusted individual authorized to manage online assets, much like an executor named in a traditional will. The executor should be entrusted with a list of one’s online financial accounts, including banking, retirement, brokerage, life insurance and payment applications as well as social media to cover one’s entire digital footprint.

A simple offline key solution can reduce cyber exposure. Instead of storing sensitive passwords or codes in cloud services or email drafts, customers should be instructed to document important accounts and access instructions in a secure offline location such as a locked drawer, safe deposit box or sealed envelope where a digital executor can find them when needed. This offline information should not be denoted as an appendix to the will, as wills may become public record after probate. The digital executor should be instructed to close accounts once transactions are settled and erase sensitive data from legacy devices before disposal.

Protecting the digital financial ecosystem

Individual preparedness, bank safeguards, and supervisory oversight all play a role in mitigating AI threats. Banks need smarter real-time defenses and practical collaboration; regulators must continue emphasizing governance, accountability and resilience; and consumers should approach digital estate planning with non-technical safeguards. Protecting digital financial assets has become both a household responsibility and a broader focus of financial system resilience. Together, these efforts can help ensure financial innovation strengthens trust within the banking industry and allows families to pass their digital financial assets across generations.