Anthropic Opens Doors to Mythos: Key Indian Entities Gain Access to Advanced Cybersecurity AI

A select group of government and private sector players in India are now utilizing Anthropic’s advanced cybersecurity model to fortify national digital infrastructure against emerging threats.

India’s digital defense landscape is undergoing a significant transformation as a handful of domestic organizations join the global cohort granted access to Anthropic’s Claude Mythos Preview. This deployment falls under the expanded scope of Project Glasswing, an initiative designed to proactively identify and mitigate high-severity software vulnerabilities before they can be exploited by malicious actors. While the identities of the participants remain confidential, sources confirm that the list includes a mix of government agencies and private sector firms, though the total count remains in the single digits for now.

Strengthening the Digital Perimeter

Project Glasswing, which debuted earlier this year, represents a shift toward using frontier models to bridge the gap in cybersecurity talent and speed. By allowing select entities to interact with the Mythos model, Anthropic is effectively crowdsourcing the defense of the world’s most critical software. In its initial phase, approximately 50 global partners successfully identified over 10,000 critical security flaws across various codebases. With the latest expansion, Anthropic has brought nearly 150 additional organizations across 15 nations into the fold, signaling an aggressive effort to scale these defensive capabilities.

Industry experts suggest that the integration of such models is timely. As software environments become increasingly complex, human-led auditing is often outpaced by the speed of discovery for zero-day vulnerabilities. By leveraging Mythos, these Indian organizations gain the ability to stress-test their own infrastructure in ways that were previously resource-intensive or technically infeasible, providing a crucial layer of protection for telecom, banking, and financial services.

Strategic Selection and Future Implications

The decision to include India in this rollout reflects the country’s growing importance in the global technology ecosystem. However, the selection process appears highly targeted; reports indicate that while cyber, telecom, and financial sectors have secured access, the broader IT services industry has largely been excluded from this specific early-access phase. This suggests a strategic prioritization of infrastructure-heavy entities that manage the backbone of the nation’s digital economy.

The expansion comes alongside broader conversations in India regarding the governance of frontier models. Regulatory bodies, including SEBI, have recently ramped up their focus on cybersecurity, forming internal task forces to oversee technical overhauls in the face of evolving digital risks. As these entities integrate Mythos, the success of the program will likely depend on how effectively these organizations can translate the AI’s findings into tangible security patches. While Anthropic has issued warnings regarding the potential for future AI systems to iterate upon their own code, the current focus remains on using these tools to secure the present, rather than theorizing about future autonomous capabilities.