OpenAI co-founder calls for AI labs to safety-test rival models

OpenAI co-founder Dario Amodei has called for AI labs to safety-test rival models, in an effort to establish a new industry standard. Amodei stated that this cross-lab safety testing would help ensure the responsible development of AI systems. OpenAI and Anthropic have already opened up their AI models for this kind of testing. The goal is to identify potential safety issues or unintended behaviors that could arise from these models, and to address them collaboratively. Amodei emphasized the importance of transparency and cooperation among AI labs, as the development of advanced AI systems carries significant risks. By sharing their models and findings, these labs hope to develop a more rigorous approach to AI safety and build public trust in the technology. The initiative is seen as a positive step towards establishing industry-wide standards and best practices for the development of safe and responsible AI systems.