Anthropic has widened public access to its cutting-edge Claude Mythos AI model, making it available to qualifying customers' security teams upon request for purposes such as vulnerability research and red-teaming. This move is expected to pave the way for widespread availability of similar models within the next 6-12 months. The expansion of access to Mythos-level models will significantly impact the capability and risk surfaces of large language models, with security implications likely to follow the hype cycle. As a result, security teams will need to reassess their strategies to mitigate potential risks associated with these powerful AI models. The increased availability of Mythos-class models will have significant security implications, making it crucial for practitioners to stay ahead of the curve1. This development matters to security professionals as it will require them to adapt their approaches to address the emerging risks associated with advanced AI models.