OpenAI and Anthropic, two prominent global pioneers in AI, have recently embarked on a groundbreaking cross-laboratory collaboration. For the past two months, they have temporarily set aside the fierce competition that often divides them to jointly scrutinize the security of their highly guarded AI models. The aim of this unprecedented partnership is to reveal any overlooked vulnerabilities in each company's internal assessments and explore potential avenues for future cooperation in ensuring AI safety and coordination. This initiative underscores the shared commitment and efforts of both companies to prioritize safety while advancing AI technology.