Trump Warms to Anthropic as Global Concerns Rise Over Powerful Mythos AI

51 articles · Updated · Axios · Apr 21

President Trump signaled renewed openness to Anthropic, suggesting a possible deal for Pentagon use of its powerful AI model, Mythos.
Mythos, deemed too dangerous for public release, is being selectively shared with major banks and tech firms due to its advanced cybersecurity capabilities.
Global regulators and finance leaders have expressed concern over Mythos’s potential for both defending and exploiting critical digital infrastructure.

If Mythos-level AI is inevitable, how can global regulators possibly keep pace with the technology's rapid advance?

Is Mythos a true cyber 'superweapon,' or just a powerful tool that exposes decades of poor security practices?

Is 'Project Glasswing' enough to secure critical infrastructure before this AI capability spreads to hostile actors?

When a private company creates a weapon-grade AI, who should have ultimate control over its use?

What stops a rival nation or actor from developing their own version of Mythos within the next year?

How does the NSA justify using a tool that its own Pentagon has officially labeled a 'supply-chain risk'?