2025 GenAI Code Security Report
AI is generating code. Our research shows it’s generating risk.
The 2025 GenAI Code Security Report analyzes the security of code generated by over 100 large language models across Java, JavaScript, Python, and C#. The results are clear: AI-generated code often isn’t secure, and the risk is likely already in your stack.
What You Will Learn:
Real Vulnerabilities, Real Impact
AI-generated code introduced risky security flaws in 45% of tests.
Which Language is Riskiest
From Java to Python, no major language was immune, but one posed the greatest risk.
Bigger Models ≠ More Secure Code
Larger, newer AI models didn’t improve security.