Consumers increasingly expect their financial institutions to use artificial intelligence–powered fraud prevention to protect them from scammers leveraging AI.
AI models trained on large financial data sets learn to spot suspicious behavior that human analysts might miss amid legitimate transactions, preventing crimes from occurring.
A survey of 2,000 U.S. adults found 97% considered fraud prevention and security the most important factors in selecting a bank, and 66% were more likely to bank with an organization employing AI security measures. Consumer choices were largely driven by the concern of 85% that AI technologies, in the hands of bad actors, made scams such as bank impersonations, voice cloning phone calls and synthetic identity fraud harder to detect, according to Alloy’s recent 2025 State of Scams Report.
"The data confirms what we're seeing on the ground: AI hasn't made fraudsters more sophisticated; it's made them more efficient," said Sara Seguin, principal adviser on fraud and identity risk at Alloy, in a statement. "But here's what's fascinating: Consumers get it. They're demanding banks use the same AI technologies to protect them."
WATCH: Artificial intelligence will drive efficiency for financial institutions in 2026.
Fighting AI-Enabled Fraud With AI
Employees can now use AI to create realistic financial documents to defraud employers, elevating the risk of insider threats within banks. Still, employee fraud comprises only a small portion of the much larger issue of financial crime.
Nilson Report estimates credit card transaction fraud will result in $43 billion in financial losses this year, but AI can enhance financial services’ security.
AI systems outpace human analysts in pattern recognition across massive data sets and can be automated and adapt faster to new types of fraud. For these reasons, banks are having analysts team with large language model–based AI assistants to query financial data sets, computer vision to analyze identity verification documents, neural networks to flag patterns of fraud across records and groups, and models to assess transaction risk.
NVIDIA launched an AI workflow for fraud detection, NVIDIA AI Blueprint, in June that improved PayPal’s real-time detection by 10% while lowering server capacity eightfold.
On the flip side, AI models must be trained on large, curated data sets of good quality to be accurate, and they don’t integrate across systems easily — raising the cost of entry for financial institutions.
Click the banner below to learn how financial services are unlocking artificial intelligence’s potential.
The Entire Finance Sector Sees the AI Need
Early AI-powered fraud detection use cases across financial services include monitoring cryptocurrency blockchain transactions for rapid fund transfers that might indicate stolen or illegal payments; chatbots capable of analyzing conversations to spot scammers; and systems designed to catch anomalous purchasing or sales behavior.
Banks are “uniquely positioned” to implement such measures due to their “central role in the payment ecosystem and access to vast amounts of historical transaction data,” according to U.S. Bank. That data can train sophisticated models for identifying fraud in real time.
Independent financial firms and advisers recognize the need to embrace AI as well. Broadridge Financial Solutions’ survey of 428 advisers found 76% believed having better technology tools would improve client acquisition.
What’s more, 51% reported adopting generative AI, although use was primarily focused on client engagement (29%) and marketing (21%).
“Client expectations are rising as investors adopt a digital-first mindset and have access to more information about financial products and investment options than ever before,” said Dale Brown, president and CEO of the Financial Services Institute, in a statement. “Across the independent financial services industry, momentum is building around the adoption of AI and other emerging technologies to increase efficiency and elevate the client experience.”
