06/04/2026
Hum
AI is rapidly becoming a frontline defense tool for critical infrastructure. 🛡️🤖
Anthropic has expanded Project Glasswing, giving approximately 150 organizations across more than 15 countries access to its powerful cybersecurity model, Claude Mythos.
The AI system was designed to identify critical software vulnerabilities and has reportedly discovered thousands of high-severity security flaws during testing.
The expanded program now includes organizations operating in sectors such as power, water, healthcare, communications, semiconductors, and national security, where a successful cyberattack could impact millions of people.
Reports indicate the initiative now spans countries including the United States, Canada, Germany, France, Japan, South Korea, India, Australia, and others, highlighting the growing role AI is playing in global cybersecurity.
As Anthropic expands Mythos, competitors like OpenAI are also advancing specialized cyber-focused AI systems, signaling a new phase in the race to secure digital infrastructure.
The real question:
Will AI become the most powerful cybersecurity tool ever created—or create new risks that defenders must learn to manage?
⚠️ Disclaimer: Access to Claude Mythos remains restricted, and reported capabilities are based on company statements and partner testing programs.