New data reveals deepfakes account for one in five biometric fraud attempts, with the rise of biometrics leading to a new cybersecurity crisis, according to research from Entrust.

Instances of deepfaked selfies increased by 58% in 2025, while other fraud tactics across biometric systems include photo of a screen, photo of a printout, 2D and 3D masks, video of a video on screen, and video of photo on screen.

This rise in deepfakes is part of a broader trend of increasingly sophisticated attacks driven by injection attacks, which surged 40% year-over-year. Injection attacks enable fraudsters to bypass live capture processes by feeding manipulated images or videos directly into verification systems.

When combined with deepfakes, these sophisticated techniques can convincingly mimic users and live capture experiences, making detection difficult without robust, multi-layered fraud prevention.

Entrust’s report also highlights trends in when fraud happens during the customer lifecycle and how it varies by industry and business models.

Sectors offering sign-up bonuses, such as cryptocurrency, face the highest rates of onboarding fraud, accounting for 67% of fraud attempts.

Meanwhile, industries with valuable, long-standing accounts, such as payments and digital-first banks, experience higher rates of account takeover (ATO) fraud, accounting for 82% and 55% of fraud attempts targeting the authentication process, respectively.

ATO fraud involves hijacking existing user accounts, typically through stolen credentials, phishing, malware, or social engineering, to steal funds or harvest sensitive data.

Building on the technical threats, the report also reveals how fraudsters are increasingly targeting individuals through psychological manipulation.

Recommended reading

• Identity Fraud Costs Soaring For Many Large Enterprises
• AI in Cybersecurity | Navigating the Promises and Risks
• 87% of Firms Hit By AI Cyber-attacks

By using tactics such as phishing, impersonation, and coercion, fraudsters are convincing victims to use their genuine identity, hand over sensitive data, or transfer funds.

The report also found that digital forgeries made up 35% of document fraud – up from a 29% average between 2022 and 2024. Overall, there’s a higher volume of physical counterfeits (47%), but typically the sophistication of digital forgeries is higher, driven by the online availability of AI tools.

“As detection improves, fraud rings evolve, becoming faster, more organised, and commercially driven,” said Simon Horswell, Senior Fraud Specialist Manager at Entrust.

“Generative AI and shared tactics fuel volumes and sophistication, targeting people, credentials, and systems.

“Identity is now the front line, and protecting it with trusted, verified identity across the customer lifecycle is essential to staying ahead of adaptive threats.”