Bybit Releases H1 2026 Security Report: Intercepts Over $700M in Potential Losses, Achieves 100% On-Chain Monitoring

Odaily News  Bybit today released its H1 2026 Risk and Security Report. Following the security incident in February 2025, Bybit has comprehensively upgraded its security architecture, transitioning toward a new defense model characterized by earlier detection, faster response, and continuous adaptation. Key highlights from the report are as follows:

User Fund Protection: In H1, over 30,000 suspicious withdrawals were intercepted, protecting nearly 20,000 users from potential losses exceeding $700 million. The average initial review time was just 4.7 minutes (with 95% completed within 10 minutes).

100% On-Chain Monitoring: Monitoring covers all business-related on-chain activities (including listed tokens, ecosystem contracts, and hot/cold wallets). In H1, 10 security incidents involving listed token projects were handled with zero platform losses; of these, responses to 8 incidents were faster than other major exchanges, and 2 attacks were detected before the project teams themselves. Additionally, approximately $212 million in potentially fraudulent on-chain funds was identified, and over 10,000 malicious addresses were blacklisted.

AI-Driven Security Operations: More than 100,000 security alerts were processed. AI-assisted audits identified critical vulnerabilities at an efficiency 3-5 times that of manual efforts; the automated red team platform reduced the time from asset discovery to initial testing to within 24 hours (compared to weeks with traditional manual methods), and the cycle from security assessment to testing was shortened from two weeks to two hours.

Accountability and Asset Recovery: In collaboration with law enforcement agencies and blockchain intelligence firms, stolen assets are being traced, and legal action has been taken against North Korea and the Lazarus Group to hold them accountable and recover funds.

David Zong, Head of Risk Control and Security at Bybit Group, stated: "The cybersecurity arms race has entered the era of 'minute-level' response. Leveraging AI to strengthen risk control capabilities and ensuring the security of AI systems themselves is our top priority, but critical security decisions remain centered on human judgment."