Businesses across North America expect AI-enabled fraud techniques to become more common through 2026, with 67 percent predicting an increase in biometric fraud, the highest expectation among surveyed categories.

Deepfake scams and synthetic identity fraud follow closely in the projections. The common thread across all three is that generative tools have lowered the cost and skill floor for attacks that previously required specialized capability. Voice cloning that once demanded substantial audio samples and technical expertise now runs from short clips through consumer applications, and document generation that supported synthetic identities has moved from manual forgery to automated production.

Biometric fraud ranking first reflects a structural problem with the authentication methods many organizations adopted specifically to replace passwords. Face and voice verification systems were designed against an assumption that generating convincing biometric samples was difficult. Generative models invalidated that assumption faster than authentication vendors updated their liveness detection, leaving a window where the newer method carries weaker protection than the one it replaced.

The financial scale sets context for the projections. Financial crime removes roughly $3.1 trillion from the global economy annually across all categories. Within the digital asset sector, Bybit's $1.5 billion breach leads a series of crypto hacks that have cost the industry billions since 2025.

Organizations responding to the forecasts have generally added verification steps outside the biometric channel rather than replacing biometric authentication outright.

Source: Visual Capitalist - https://www.visualcapitalist.com/sp/ig02-8-predictions-for-the-future-of-fraud/