Analyzing Hardware Designs Generated by Large Language Models for Vulnerabilities

Around 60% of hardware designs generated by LLMs are prone to CWEs, emphasizing the need for caution and thorough verification processes.