Anthropic has officially launched Claude Fable 5, a 'Mythos-class' AI model designed with extensive safeguards to mitigate potential misuse, particularly in cybersecurity. Initially, Anthropic refrained from releasing its first Mythos model due to concerns about its hacking capabilities, opting instead to share it with a limited group of companies and the US government to enhance cybersecurity measures. The new model, while powerful, comes with restrictions that direct certain queries to Anthropic's Opus 4.8 model. Anthropic emphasized that Fable 5 surpasses previous models in performance across various benchmarks, including software engineering and scientific research. Notably, Stripe found that Fable 5 could condense months of engineering tasks into mere days, while IMC noted its exceptional performance in trading analysis. Furthermore, the release aligns with Anthropic's strategic move towards an IPO, as it aims to expand access to its AI models through a trusted access program. This launch occurs amidst heightened scrutiny from governments regarding AI capabilities and security implications.