Cloud Hypervisor launched a policy prohibiting artificial intelligence (AI)-generated code contributions, but project contributors warn the rules will likely be broken from day one. Many developers now use AI tools for code refinement, suggestions, or debugging without considering such help as “AI-generated” content. Security concerns fuel much of the caution around AI-generated code. A 2025 Veracode study found that 45% of all AI-generated code contained vulnerabilities, with common problems including weak defenses against cross-site scripting and log injection attacks. Research also shows that trying to fix AI code can make problems worse.