CertiK launches AI Auditor: Balancing low false positives and high accuracy, real-world testing shows it identifies 88.6% of real vulnerabilities.

CoinFeed reported on April 7th that Web3 security company CertiK announced the launch of its AI auditing tool, AI Auditor, and simultaneously released a series of open-source integration solutions for AI Coding Agents. Originally developed as an auxiliary tool for its internal security experts, the system has been officially released to the public after more than six months of real-world scenario validation. Reportedly, in an assessment of 35 real Web3 security incidents in 2026 (test data not used for model training or knowledge base construction), CertiK AI Auditor accurately identified the root causes of 88.6% of the vulnerabilities. The system is designed with a high signal-to-noise ratio, significantly reducing false positives while maintaining high detection capabilities.