India Among Countries Granted Access To Anthropic’s Advanced Cyber Security AI Model

Artificial intelligence company Anthropic has expanded international access to its cyber security-focused AI model, Mythos, bringing India into a select group of countries participating in its Project Glasswing programme.

Invest in Indian Markets and Unlock Future Potential With 5paisa!

Open Account Now

The move broadens the reach of the technology beyond its initial rollout in the United States and the United Kingdom. According to reports, around 150 organisations across more than 15 countries will now gain access to the AI model, which is designed to identify and help address software security vulnerabilities.

Anthropic had previously restricted access to Mythos, citing concerns that its advanced capabilities could potentially be misused if released without safeguards.

Project Glasswing Expansion

The latest expansion includes countries such as India, Canada, Australia, New Zealand, France, Germany, Italy, Switzerland, the Netherlands, Spain, Belgium, Sweden, Japan and South Korea, as per reports. The programme has also been extended to the North Atlantic Treaty Organisation (NATO).

Project Glasswing was launched to support organisations responsible for critical systems and infrastructure by helping them detect and address cyber security weaknesses before they can be exploited.

Anthropic stated that participating organisations were selected based on the potential impact of a successful cyber attack on their systems.

Organisations Included In The Programme

The expanded list of participants includes companies operating in technology, cyber security, financial infrastructure and communications.

Among the organisations granted access are South Korean technology firms Samsung Electronics, SK Hynix and SK Telecom, along with identity management company Okta.

Financial infrastructure providers including Euroclear, Intercontinental Exchange and SWIFT have also joined the programme. The European Union Agency for Cybersecurity (ENISA) is among the institutions included in the latest phase.

Anthropic said the initiative now covers industries that had limited representation during the initial rollout, including healthcare, power, water, communications and hardware.

Focus On Critical Infrastructure

According to the company, the programme is intended for organisations whose systems play an important role in public services, national infrastructure and economic activity.

Anthropic noted that a major cyber attack affecting many of the participating organisations could potentially impact more than 100 million people, highlighting the importance of strengthening cyber resilience across essential sectors.

The company also indicated that future expansion of Project Glasswing could include additional infrastructure operators, maintainers of critical open-source software and organisations involved in cyber safety testing.

Wider Availability After Limited Rollout

Anthropic introduced Claude Mythos Preview in April with access limited to roughly 50 organisations, most of them based in the U.S. The restricted rollout reflected concerns around the model’s advanced coding and vulnerability-detection capabilities.

Before this latest expansion, access had already been provided to technology companies including Microsoft, Apple and Oracle, as well as financial institution JPMorgan Chase and cyber security firm CrowdStrike.

With India now included in the programme, organisations operating critical systems will gain access to one of the most closely controlled AI tools currently being deployed in the cyber security space.