Bybit Protects Users from $300M in Fraud with AI-Driven Defense Framework

Bybit, a leading cryptocurrency exchange, has made significant strides in preventing fraud and scams through its AI-powered risk framework. In the fourth quarter of 2025, Bybit's system intercepted $300 million in fraudulent withdrawals, safeguarding over 4,000 users.

The exchange's defense strategy consists of a three-tier framework that detects unusual patterns, flags suspicious accounts, and blocks high-risk transactions. At the core of this approach is a risk-based dynamic protection system that classifies fraud scenarios into three categories: low, medium, and high.

Bybit's AI algorithms identified 350 high-risk addresses through on-chain monitoring, allowing it to safeguard an additional 8,000 users. The exchange also worked with external monitoring partners, including TRM, Elliptic, and Chainalysis, to block over 3 million credential theft attempts throughout the year.

The global cryptocurrency industry saw $17 billion lost to fraud in 2025, according to Chainalysis estimates. Bybit's Group Risk Control Director emphasized that standardized monitoring indicators are shared across the crypto ecosystem, promoting a safer environment for all users.