Checkmarx One Hybrid Engine Triples Industry F1 Score, Slashes False Positives by 60%
Event summary
- Checkmarx One’s new hybrid SAST scanning engine achieves F1 score of 0.64, 3x higher than competitors' average of 0.20.
- Engine reduces false positives by 60% while extending coverage to AI-generated and emerging languages.
- Finding Analysis Engine (FAE) introduced to confirm true positives and suppress false ones.
- Early access available now; virtual summit Agentic AppSec Unleashed ’26 scheduled for June 16, 2026.
The big picture
The surge in AI-generated code has created a new vulnerability landscape, with 49% of production code now AI-generated and measurably less secure. Checkmarx’s hybrid engine addresses this by combining deterministic rules with AI-assisted scanning, offering a solution that balances precision and coverage. This development reflects a broader industry shift toward integrating AI into security tools while maintaining governance and risk management standards.
What we're watching
- Adoption Pace
- How quickly enterprises will integrate hybrid scanning engines to address AI-generated code vulnerabilities.
- Competitive Response
- Whether legacy tools and AI-focused competitors can match Checkmarx’s F1 score and false positive reduction.
- Market Differentiation
- The extent to which Checkmarx can sustain its lead in scanning fidelity as AI-assisted development evolves.
Related topics