Generative AI coding assistants like GitHub Copilot amplify existing security issues in codebases by replicating vulnerabilities and bad practices, even if the original code is already secure. The tool's behavior can exacerbate security problems, reinforcing bad habits, overlooking security concerns, and introducing outdated or flawed patterns. To mitigate this issue, organizations should conduct manual reviews of AI-generated code, implement SAST guardrails, adhere to secure coding guidelines, provide training and awareness to development teams, prioritize and triage issues, and consider mandating security guardrails for the use of generative AI code assistants. By combining these techniques with traditional AppSec methods, developers can strike a balance between innovation and security, making their applications more resilient to potential threats.