最新报道:Cybersecurity researchers have identified a new "CopyPasta License Attack" method that exploits AI coding assistants by embedding hidden prompt injections in common developer files like LICENSE.txt and README.md. This causes AI systems to automatically replicate malicious code across projects without detection. The attack relies on developers delegating documentation tasks to AI tools, bypassing traditional human review processes. While requiring user interaction to propagate, it highlights growing risks as AI adoption increases in software development.