OpenAI and Anthropic have initiated a protocol for cross-lab safety testing of AI models to establish industry standards. This protocol seeks to enhance collaboration among AI labs, addressing security vulnerabilities through shared assessments and rigorous methodologies, thereby mitigating risks associated with AI misusage across various applications and organizations globally.
🔧 OpenAI Advocates for Cross-Lab AI Safety Testing
