NSE:HAL

Stock Market Today

  • Australia Gains Access to Anthropic's Highly Sensitive AI Cybersecurity Model
    June 2, 2026, 8:40 PM EDT. Australian organisations are now granted access to Anthropic's AI model Claude Mythos, deemed "too dangerous" for public release due to its powerful cybersecurity capabilities. The model excels at identifying critical software vulnerabilities that could otherwise be exploited to trigger major breaches. This access is part of Project Glasswing, a staggered global rollout that includes over 150 organisations across 15 countries, covering sectors like finance, power, water, healthcare and communications. Anthropic has involved key Australian institutions such as the Treasury and Reserve Bank. The program has already uncovered over 10,000 severe vulnerabilities since April, aiming to strengthen protection of critical infrastructure ahead of broader release.