Anthropic’s Mythos controversy puts AI export controls under pressure

WIRED’s reporting on SK Telecom and Anthropic’s Claude Mythos controversy shows how complicated AI access controls are becoming as frontier models move through partner programs.

The issue centers on whether commercial relationships and telecom infrastructure create unacceptable exposure when advanced AI systems are involved. That makes model access a national-security question, not only a product licensing decision.

For AI companies, the episode points to a harder compliance environment where partner vetting, export rules, and geopolitical risk may shape who can use frontier systems.