NSA uses Anthropic's advanced Mythos model to attack networks in China and Iran. Cooperation continues despite the start-up being listed as a threat.
OpenSquilla has released the MetaSkill framework, which aims to give AI agents more control over their own processes for automating complex tasks. Meanwhile, the NSA is reportedly using Anthropic's advanced Mythos model for cyber operations targeting China and Iran, a collaboration that continues despite the startup's inclusion on a national security threat list. AI
IMPACT New AI agent control frameworks could enable more sophisticated automation, while government use of advanced models highlights evolving cyber warfare capabilities.