Bybit Earns $300M for Thousands of Users With AI-Enhanced Fraud Prevention: Report

Beyond the acquisition, Bybit blocked attempts to install 3 million credentials in line with plans to take over accounts by 2025.

Bybit reported receiving $300 million from thousands of users at a time when crypto-related fraud remains at an all-time high across the industry.

The exchange revealed that these attempts were exposed to an AI-powered fraud detection system that intervenes before people lose their money.

Consequences of Protective Measures

Bybit shared the results of the 2025 Security Initiative, saying on social media,

“We raised the bar by 2025, catching $300M in impersonation and fraud with our new AI-driven risk framework.”

The announcement comes as crypto fraud continues to weigh on the sector, with data from Chainalysis showing that 17 billion dollars in digital assets were lost in fraud and fraud cases by 2025.

The report reveals that in the fourth quarter alone, the exchange flagged $500 million in withdrawals for revisions. Of that amount, $300 million was successfully captured and recovered, protecting the savings of more than 4,000 users.

At the same time, Bybit’s proprietary AI models identified 350 high-risk fraud addresses using on-chain data, protecting 8,000 people from potential withdrawal losses. The company also reported that it blocked more than 3 million hacking attempts linked to account takeover attempts by 2025.

In addition, its system automatically recorded 350 suspicious addresses, and included 600 more through internal ticketing, preventing another $1 million in fraud losses.

You may also like:

David Zong, Head of Group Risk Control at Bybit, said in a statement that the company’s goal in 2025 was to transform risk management into an effective and intelligent guardian by integrating AI and on-chain monitoring.

“By combining AI-driven monitoring with real-time intelligence from industry partners such as TRM, Elliptic and Chainalysis, we not only protect Bybit users but also help map the DNA of fraud networks,” he wrote.

A Three Tier Risk Framework

Bybit’s security model creates three-stage fraud scenarios while maintaining normal trading activity. At a low risk level, the platform uses big data analysis to detect unusual activity, such as mass withdrawals from newly created addresses, and uses automated surveys to support its Risk Operations team in blocking suspicious sites.

Real-time alerts are triggered during the withdrawal process for medium-risk cases, such as accounts flagged by verification databases or linked to questionable withdrawal addresses. This, in turn, encourages people to review activities that may be influenced by civil engineering techniques.

At the highest level, wallet addresses associated with confirmed scams face an immediate withdrawal ban and a mandatory one-hour cooling-off period.

The report concludes by outlining common monitoring indicators for industry-wide implementation, including a proof-of-concept stuffing engine, real-time AI pattern recognition in pig slaughter, an integrated intelligence hub that includes tools from TRM Labs, Elliptic, and Chainalysis, and an end-to-end bag tracking model for illegal tracking.