News

Bybit Recovers $300M Using AI-Powered Crypto Fraud Detection

By

Shweta Chakrawarty

Shweta Chakrawarty

Bybit’s new AI-driven risk framework intercepted $300 million in fraudulent withdrawals in Q4 2025, protecting 4,000 users.

Bybit Recovers $300M Using AI-Powered Crypto Fraud Detection

Quick Take

Summary is AI generated, newsroom reviewed.

  • Bybit’s AI system intercepts $300 million in suspicious withdrawal attempts.

  • Platform flags $500 million in total high-risk activity during Q4 2025.

  • Security initiative identifies 350 high-risk addresses with help from Chainalysis.

  • Proactive cooling-off periods prevented losses for over 4,000 retail users.

Crypto exchange Bybit says it has recovered more than $300 million in scam-related funds using its new AI security system. The results come from the platform’s 2025 security initiative, shared in late February 2026. According to the company, its tools flagged about $500 million in suspicious withdrawals during Q4 2025 alone. 

From that pool, roughly $300 million was intercepted before leaving the platform. Bybit said the effort directly protected more than 4,000 users. It also helped shield thousands more. The move comes as crypto fraud continues to rise sharply across the industry.

Surging Crypto Fraud Landscape

The broader crypto space is facing a wave of increasingly complex scams. Industry data shows billions of dollars were lost to fraud in 2025. Through AI-powered impersonation scams have exploded. With some estimates pointing to a massive year-over-year surge. These scams often use deepfakes, fake trading platforms and coordinated wallet networks. As a result, exchanges are under growing pressure to detect threats earlier. Bybit’s latest push clearly aims to stay ahead of this curve rather than react after losses occur.

Bybit’s AI-Powered Defense Framework

At the center of the effort is Bybit’s multi-layer risk system. The exchange says the platform monitors on-chain activity in real time using AI models. That trained to spot abnormal wallet behavior. When the system detects suspicious patterns. It can automatically flag or delay withdrawals. During the latest review period, the system identified about 350 high-risk addresses linked to fraud schemes. 

Bybit built the framework with help from blockchain intelligence firms. Including TRM Labs, Elliptic and Chainalysis. The exchange says the approach focuses on prevention first. In some cases, withdrawals enter a cooling-off period. In others, assets can be frozen while risk teams investigate. The goal is simple: stop scams before funds disappear into complex laundering networks.

Q4 2025 Results Show Early Impact

The numbers from the fourth quarter highlight how aggressive the new system has become. Out of roughly $500 million in flagged suspicious withdrawals. About $300 million is successfully intercepted or recovered. Bybit said more than 4,000 users were directly protected during this period. In addition, up to 8,000 users may have avoided losses through early address blocklisting and warning systems. The exchange framed this as a shift from reactive security to proactive defense. Instead of chasing stolen funds after the fact, the platform aims to block bad activity at the door.

Raising the Bar for Exchange Security

Bybit believes this initiative sets a higher standard for crypto platform safety. As fraud tactics evolve, exchanges are increasingly turning to AI. Additionally, real-time analytics to keep users safe. Still, the fight is far from over. Scam networks continue to adapt quickly. But the exchange’s latest results show that stronger monitoring and industry partnerships can make a real difference. If similar systems spread across the sector, everyday crypto users may finally see a safer trading environment ahead.

Google News Icon

Follow us on Google News

Get the latest crypto insights and updates.

Follow