AI can write your code, but nearly half of it may be insecure - Help Net Security

Veracode's 2025 GenAI Code Security Report reveals that code generated by Large Language Models (LLMs) contains security vulnerabilities in 45% of cases, presenting weaknesses aligned with the OWASP Top 10. Specific issues include an 86% failure rate to defend against cross-site scripting (CWE-80) and 88% vulnerability to log injection attacks (CWE-117).