AI with excessively high cyberattack capabilities, "Claude Mythos Preview," can develop attacks from already known vulnerabilities "N-day" within hours, leading to a paradigm shift from "N-day to N-hour," points out Anthropic – GIGAZINE https://www.yayafa.com/2818860/ # AgenticA
Reports indicate the U.S. National Security Agency (NSA) is preparing to use Anthropic's Mythos AI model for offensive cyber operations. This comes after the Department of Defense previously identified Anthropic as a supply chain risk. Additionally, Anthropic is reportedly providing access to Mythos, a model focused on cybersecurity, to the European Union's cyber agency and Australia. AI
IMPACT Potential for advanced AI-driven cyberattacks and defense strategies, raising geopolitical and security concerns.