AI deepfakes used in financial fraud and identity abuse affecting financial sector
Deepfake technology has enabled rising financial fraud and identity abuse targeting businesses and individuals. A Hong Kong engineering firm lost $25 million through deepfake video impersonation, and 92 percent of financial sector organisations reported being targeted by deepfake fraud in 2024. Deepfakes have also been weaponised for reputational and sexual abuse, with women making up 99 percent of victims of non-consensual deepfake pornography.
What happened
Deepfake technology has been increasingly used to commit financial fraud and identity abuse. A global engineering firm Arup lost $25 million to criminals using deepfake video impersonation of company executives on video calls. An agency leader at WPP was nearly deceived by a deepfake video they believed was CEO Mark Read during a Teams meeting.
In 2023, deepfake fraud incidents in the fintech sector rose by 700 percent. The average loss per fintech company reached $637,000 in 2024, with nearly a quarter of firms losing over $1 million. Overall, 92 percent of financial sector organisations reported being targeted by deepfake fraud in 2024, and 10 percent lost over $1 million.
Deepfakes have also been weaponised for reputational and sexual harm. Northern Irish politician Cara Hunter was targeted with a deepfake pornographic video during an election campaign. Studies show that 90-95 percent of deepfakes online are pornographic, with 99 percent of victims being women.
Reported impact
Financial losses from deepfake fraud are substantial and rising. Deloitte forecasts fraud losses from deepfake incidents to reach $40 billion in the United States by 2027.
The psychological and reputational damage to victims can be devastating, particularly in cases involving non-consensual deepfake pornography. The scale of the problem is vast, with criminals using deepfake technology across finance, social media, and other sectors.
Index analysis
Deepfake-enabled financial fraud is prosecuted under existing criminal fraud and cybercrime statutes in nearly every country. However, the dramatic rise in such scams has prompted banks and fintechs to invest in detection and prevention.
Denmark has proposed a 2025 law designed to grant individuals copyright-like rights over their own face, body, and voice. The proposed law addresses privacy, reputational, and commercial harms through notice and take-down procedures, with platforms facing fines for non-compliance. The US has proposed the NO FAKES Act to give people more control over the use of their image and voice.
Governance controls
The AI community is developing guardrails including digital signatures and watermarks to identify deepfake content. Deepfake detection is emerging as a field holding promise to identify if digital content was generated by AI, though scammers can potentially circumvent these tools.
Existing criminal fraud and cybercrime statutes are used to prosecute deepfake-enabled financial fraud, though these may not address all harms. Platforms are implementing detection and prevention investments to combat the rising threat.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.