Experts Warn AI Vulnerability Threat Pre-Dates Anthropic's Mythos Release

Story Summary

Cybersecurity experts state that vulnerabilities revealed by Anthropic's Mythos model can be detected using existing AI tools, contradicting claims of a novel threat. While global institutions scramble to address potential risks, researchers note that the offensive capability to find flaws predates the announcement, leaving systems exposed because defensive measures lag behind. This dynamic accelerates attack frequency despite ongoing efforts by major firms to develop countermeasures.