Putting an end to the AI cyber responsibility turf wars

The article argues that AI security must be treated as a shared, continuous responsibility across regulators, AI model developers, and organizations using AI, urging comprehensive legislation, secure-by-design development practices, and organizational safeguards such as risk assessments and red-teaming. It emphasizes that current debate focuses too much on ethics while insufficiently addressing concrete cybersecurity risks from LLMs (e.g., data poisoning, model evasion, malware misuse), and calls for coordinated action—regulatory frameworks, developer oversight, and implementation-level controls—to minimize the expanding attack surface as AI adoption accelerates.