NSA using Claude Mythos for 'offensive cyber operations,' report claims

A recent report claims that the U.S. National Security Agency (NSA) is using Anthropic’s cybersecurity-focused AI model, Mythos, for 'offensive cyber operations,' with the help of several embedded eng... Read original →

The NSA’s deployment of Mythos for offensive cyber operations reveals the operational maturity of AI in automating zero-day discovery and exploit workflows. This will accelerate adoption of AI-optimized secure enclaves in semiconductor designs, particularly RISC-V-based TEEs. The DoD’s prior blacklisting of Anthropic—stemming from its refusal to enable autonomous weapons—exposes deep fissures within U.S. national security policy on AI ethics. Competitors like OpenAI may rush compliant, militarized model variants to defense contracts, but risk triggering tighter export controls under emerging AI governance regimes. If institutionalized, the Trump-era executive order mandating pre-release model handover could force U.S. AI firms into a national-security-first development paradigm akin to semiconductor export restrictions, ultimately eroding their global commercial edge.