- CertiK has launched its AI Auditor publicly after using it internally for more than six months.
- The company said the system reached an 88.6% cumulative exact hit rate across 35 real-world Web3 security incidents from 2026.
CertiK is pushing further into AI-based security tooling, this time with a product aimed directly at developers, auditors and institutions trying to catch vulnerabilities before they become losses onchain.
The New York-based Web3 security firm said its newly released AI Auditor is designed to deliver what it called low-noise, high-signal intelligence, with the goal of helping teams identify serious issues earlier in the development cycle. The company said the tool had been used internally for over six months before being opened to the public.
A security tool built to cut through alert fatigue
According to CertiK, the system was tested against 35 real-world Web3 security incidents from 2026 and achieved an 88.6% cumulative exact hit rate. The company framed that result not just as a detection metric, but as evidence that AI tools can be made more usable for security teams if they produce fewer irrelevant alerts.
That point matters. One of the more persistent complaints around AI-assisted security products is that they often generate too much noise, forcing engineers to spend time sorting weak or duplicated findings rather than addressing real risk.
CertiK said its model addresses that through a MultiScanner framework, which runs specialized scanners in parallel, and a proprietary Multi-Stage Validator that filters, deduplicates and assesses findings for semantic validity and exploitability before surfacing them to users.
Security moves closer to the developer workflow
The company is also releasing open-source integrations for AI coding agents, a sign of where this is heading. Rather than treating security as a separate review step at the end of development, CertiK is trying to insert it directly into everyday workflows.
It also said the AI Auditor draws on a continuously updated knowledge base containing exploit data, audit findings and attack patterns, allowing the system to reflect current threat intelligence rather than relying only on static model training.
For now, CertiK is positioning the tool as a complement to human auditors, not a replacement. The software handles baseline detection, pre-audit triage and continuous monitoring, while human reviewers remain focused on deeper protocol risks and more complex vulnerabilities.