Search
As reported by SiliconAngle, Anthropic's content management system inadvertently revealed its new-generation, large-scale language model, Claude Mythos—currently in the testing phase—due to a configuration error. This model surpasses the performance of the existing premier model, Opus 4.6, especially in areas such as programming, academic reasoning, and cybersecurity. Documents that were leaked indicate that Mythos will introduce a new, fourth-tier product, Capybara, which is set at a higher tier and entails greater operational expenses. Given the potential cybersecurity threats, Anthropic has opted for a cautious rollout, prioritizing access for defense organizations to fortify its code. This development has sparked market anxieties regarding AI-fueled attacks, resulting in a drop in the stock values of cybersecurity behemoths like CrowdStrike and Palo Alto Networks.
-
C114 Communication Network -
Communication Home
