Using GEPA to detect malicious AI-generated code
A publication dated October 5, 2025 reports that prompts found with GEPA detected malicious behavior in AI-generated code, blocking 90% of malicious submissions with 1% of the audit budget.
Source: Prompts com GEPA para Detectar Código Malicioso Gerado por IA (x.com). Text prepared with AI from this source.
What happened and what to do
On October 5, 2025, a publication described using prompt optimization with GEPA to assess AI-generated code. According to the report, the prompts detected malicious behavior and blocked 90% of malicious submissions using 1% of the audit budget.
A company could add a security analysis stage to its AI-generated code workflow: collect submissions, run automated checks, route suspicious cases for review, and track detections and blocks in a dashboard. The process should be tested with the company’s own data and criteria set by its security team.
How the consultancy can help
Wendelmaques can diagnose the company’s risks and code-generation workflow, scope an analysis solution, and implement and operate collection, automated checks, and monitoring.
Next step
Send a short description of your case to receive a scoped proposal for diagnosis and implementation.
Let's assess your project
I reply with scope and a proposal.