Reports revealed that a US security agency is deploying Anthropic's Mythos model internally, despite Pentagon attempts to blacklist the company from certain government contracts. This comes amid growing concerns about the model's exploratory capabilities to exploit security vulnerabilities in financial systems.

This case highlights the growing tension between the need for advanced AI tools for defense purposes and concerns about governance and oversight of frontier models. On one hand, security agencies seek to benefit from the latest technologies to confront cyber threats, and on the other hand, regulatory bodies fear these models could be used to destabilize financial systems.

This situation is an indicator of gaps in government oversight frameworks, where agencies sometimes act independently away from central directives. This controversy is expected to accelerate discussions on export controls and dual-use AI regulation.