AI Deepfakes Drove 40% of High-Value Crypto Fraud Last Year: Report

AI-generated deepfake impersonations of presidency officers, billionaires, and celebrities accounted for 40% of “high-value frauds” in 2024, a Bitget report revealed.

That very same 12 months, $4.6 billion in crypto was misplaced to scams, a 24% enhance from the earlier 12 months, Bitget’s Anti-Rip-off Report 2025—co-authored with Slowmist and Elliptic—discovered.

“Crypto scams have entered a brand new period—pushed by AI deepfakes, social engineering, and misleading venture fronts,” the report stated. “Scams now exploit belief and psychology as a lot as expertise. From pockets takeovers to multimillion-dollar frauds, the assaults have gotten extra personalised, extra plausible, and tougher to detect.”

A “frequent” deepfake that appeared, in keeping with the report, featured Tesla CEO Elon Musk pitching fraudulent funding or giveaway schemes. Different makes use of for deepfakes embody bypassing know-your-customer verification, creating digital identities to steer funding scams, and conducting faux Zoom phishing assaults.



The Zoom rip-off sees scammers impersonate executives, consultants, and journalists—together with Decrypt reporters—to get a sufferer on a fraudulent video name. This may increasingly see the attacker provide the sufferer a job or ask to interview them for a non-existent article. In the course of the faux Zoom name, attackers can acquire management of the sufferer’s pc, steal information, and doubtlessly entry personal crypto keys. 

Bitget reviews that in some circumstances, the attackers are utilizing deepfake instruments to generate video and audio content material to trick the sufferer into becoming a member of the decision.

Many of those deepfake scams aren’t new; the Elon Musk rip-off first went viral in 2022. Nonetheless, with the speedy acceleration of synthetic intelligence, deepfakes have began to look actual. A lot in order that President Trump signed the bipartisan Take It Down Act final month that protected the victims of deepfake porn—though, deepfakes typically nonetheless aren’t prohibited. 

In Could, actor Jamie Lee Curtis referred to as out Meta CEO Mark Zuckerberg after discovering a faux AI-generated advert of herself selling a product with out her consent.

“The largest risk to crypto at this time isn’t volatility—it’s deception,” Bitget CEO Gracy Chen stated, in a launch. “AI has made scams quicker, cheaper, and tougher to detect.”

Exterior of deepfakes, social engineering and fashionable Ponzi schemes ranked second and third because the report’s “most harmful” scams. 

By definition, social engineering scams exploit the psychology of victims in a technique that’s “low-tech, but extremely efficient,” the report says. Probably the most frequent examples is the pig butcher rip-off, additionally referred to as a romance rip-off, which sees the attacker develop a relationship with the sufferer solely to rip-off them.

As for the ol’ traditional Ponzi scheme—named after the early twentieth century scammer Charles Ponzi—the report claims that the rip-off has gone by way of a “digital evolution.”

“These scams sometimes disguise themselves in new ideas like DeFi, NFTs, and GameFi, packaged as venture fundraising, liquidity mining, or platform token staking,” the report states. “Basically, they continue to be traditional Ponzi schemes the place ‘new cash fills previous holes.’ As soon as the money circulation breaks or the operators money out and exit, the whole system collapses shortly.”

The report notes an increase in Ponzi schemes that undertake “game-like person interfaces” and use deepfakes to forge endorsements from celebrities to spice up credibility within the scheme. AI has revolutionized the scamming trade, leading to a markedly totally different panorama from only a few years in the past.

“5 years in the past, avoiding scams meant ‘don’t click on suspicious hyperlinks,'” the report completed. “At present, it’s ‘don’t belief your individual eyes.'”

Edited by James Rubin