Binance founder Changpeng Zhao (CZ) has sounded a fresh alarm on the growing threat of AI-powered deepfakes, after a chilling real-life hack compromised Japanese crypto advocate Mai Fujimoto during a routine video call.

The incident has renewed calls for stricter security habits as criminals wield advanced AI to breach even trusted communication channels.

A Case Study in Modern Hacking

On June 20, Fujimoto, better known online as Miss Bitcoin, narrated how she fell victim to a deepfake after being tricked into a Zoom call with an acquaintance whose Telegram account had already been compromised.

According to her, the person on the video link was familiar, causing her to let down her guard. “For about 10 minutes in the online meeting, I saw her face but had no clue it was a deepfake,” Fujimoto recounted.

However, there were audio issues with the call, prompting the deepfake impersonator to send a link supposedly meant to fix the problem. After clicking the link, Fujimoto unknowingly installed malware that compromised her Telegram and MetaMask. Her main X account was also hijacked, forcing her to plead with followers to report it as impersonation. She also warned users of the social platform not to click any links sent from the account.

“If I had known about this kind of attack, I might not have clicked the link,” the crypto enthusiast lamented, urging widespread awareness.

Her ordeal struck a chord across the crypto community, with influencers and security experts flooding X with stern reminders that in the age of AI, trust must be verified through multiple channels.

CZ weighed in, too, warning his 10 million followers of the rampant use of artificial intelligence in new kinds of deepfake hackings. “Even a video call verification will soon be out of the window,” he cautioned, suggesting AI-generated impersonations will become harder to detect as the technology improves.

The former Binance head urged users never to install software from unofficial links, even if they appear to come from friends.

The Rise of AI-Powered Fraud

The attack on Fujimoto is hardly isolated. A major report released by Bitget only days ago revealed that deepfake technology had played a role in nearly 40% of all high-value crypto frauds in 2024, contributing to at least $4.6 billion lost to scams.

The study documented criminals using AI to create convincing fake videos of public figures like Elon Musk promoting scams, simulating customer service chats, and crucially, weaponizing video conferencing tools like Zoom with malicious links.

Additionally, a different publication by Chainalysis noted that lawbreakers were increasingly using AI tools, including deepfakes, to bypass KYC measures and automate fraud, making crypto cons “increasingly difficult to detect.”

In one such instance, Hong Kong police arrested 31 individuals linked to a syndicate that stole $34 million using AI-generated videos of crypto executives. Terrifyingly, the report warned that AI-driven swindles are poised to grow significantly.

---

Sandeep Nailwal, co-founder of Polygon, has flagged a troubling rise in deepfake scams using his likeness to deceive crypto founders. Polygon is one of the largest Ethereum layer-2 networks.

In a May 13 social media post on X, Nailwal said several individuals recently reached out to verify if he had spoken with them over Zoom.

However, he was not involved in those meetings as the scammers used manipulated videos of him to gain trust and trick them into installing harmful software.

How the scammers operate

Nailwal stated that the scam began with the compromise of the Telegram account belonging to Shreyansh Singh, who leads Polygon Ventures.

From there, the attackers messaged startup founders affiliated with Polygon’s investment network, pretending to reconnect for funding discussions.

Interested recipients were then invited to join Zoom meetings through phishing links disguised to look official.

These links required desktop access and led to video calls featuring AI-generated versions of Nailwal, Singh, and a woman claiming to be part of the investment team.

Nailwal pointed out that the calls did not have audio. Instead, participants were urged to install a Software Development Kit (SDK).

However, the Polygon founder said the move is designed to infect the recipient’s systems.

Rising concerns about deepfakes

Nailwal’s incident is not an isolated case and shows that deepfake fraud is becoming a more dangerous industry-wide issue.

Over the years, deepfake impersonations have increasingly targeted prominent crypto executives like Ripple CEO Brad Garlinghouse, who has also faced multiple instances of his deepfakes being used in scam promotions.

This attack vector shows cybercriminals’ growing use of generative AI in the evolution of crypto-related fraud. A recent report pointed out that losses from deepfake-driven scams exceeded $200 million in Q1 2025 alone.

Due to this, Nailwal advised community members never to install unfamiliar software during unsolicited interactions.

He also emphasized the importance of operational hygiene, urging crypto users to separate their wallet-signing activities from general device use. He stated:

“These attacks keep getting more and more sophisticated, so best approach is to keep a separate laptop for signing via your wallets only from that laptop and never do anything else on that wallet.”

Mentioned in this article

---

Changpeng “CZ” Zhao, former CEO of Binance, stated Thursday that he had encountered an AI-generated video that replicated his voice so accurately that he could not distinguish it from a real recording.

The post, shared via X, featured an AI voice-over of Zhao speaking Mandarin, matching his facial movements with a precision he described as “scary.”

The video shows Zhao speaking in Chinese over a series of video clips, AI-generated content, and photos. Its fidelity renewed concerns around the unauthorized use of AI to impersonate public figures.

The incident adds to a growing body of cases where digital likenesses of crypto executives have been cloned using generative tools, sometimes for fraudulent purposes.

Zhao, who remains a prominent figure in the crypto industry despite stepping down as Binance CEO in 2023, has previously issued warnings about impersonation attempts involving deepfakes. In October 2024, he advised users not to trust any video footage requesting crypto transfers, acknowledging the circulation of altered content bearing his likeness.

Deepfakes and crypto: Increasing operational risk

The latest video adds a new dimension to impersonation tactics that have moved beyond static images and text. In 2023, Binance’s then-Chief Communications Officer, Patrick Hillmann, disclosed that scammers used a video simulation of him to conduct meetings with project representatives via Zoom.

The synthetic footage was stitched together using years of public interviews and online appearances, enabling actors to schedule live calls with targets under the pretense of official exchange engagement.

Zhao’s experience suggests voice replication has reached a comparable level of realism, even to the person being mimicked, raising fraud risks beyond social media impersonation.

In February, Arup’s Hong Kong office staff were deceived into transferring approximately $25 million during a Microsoft Teams meeting, believing they were speaking with their UK-based finance director. According to the South China Morning Post, every participant on the call was an AI-generated simulation.

Voice-cloning capabilities now require minimal input

Tools once dependent on extensive voice samples now operate with only brief recordings. Many consumer-level systems, such as ElevenLabs, require less than 60 seconds of audio to generate a convincing clone. The financial institution reported in January that over one-quarter of UK adults believe they encountered scams involving cloned voices within the prior 12 months.

These tools are increasingly available at low cost. According to threat intelligence briefings from CyFlare, turnkey access to voice-to-voice cloning APIs can be purchased for as little as $5 on darknet marketplaces. While commercial models offer watermarking and opt-in requirements, open-source and black-market alternatives rarely adhere to such standards.

The European Union’s Artificial Intelligence Act, formally adopted in March 2024, includes mandates that deepfake content must be clearly labeled when deployed in public settings. However, the law’s enforcement window remains distant, with full compliance not expected until 2026.

Without active regulatory barriers, hardware manufacturers are beginning to integrate detection capabilities directly into consumer devices.

Mobile World Congress 2025 in Barcelona featured several demonstrations of on-device tools designed to detect audio or visual manipulation in real-time. While not yet commercially available, these implementations aim to reduce user dependence on external verification services.

Mentioned in this article

---