Daijiworld Media Network - New Delhi
New Delhi, Jun 3: India will be among the countries gaining access to Anthropic’s advanced cybersecurity-focused artificial intelligence model, Mythos, as the company expands its Project Glasswing initiative to around 150 organisations across more than 15 countries, according to a report by the Financial Times.
The expanded programme will provide access to organisations in countries including Canada, Australia and New Zealand, which are part of the Five Eyes intelligence alliance, as well as major economies in Europe and Asia such as France, Germany, Italy, Switzerland, the Netherlands, Spain, Belgium, Sweden, Japan, South Korea and India.
In a statement issued on Tuesday, Anthropic said it is extending the partnership to approximately 150 new organisations. The company noted that all participating organisations will be required to meet its security standards before being granted access.
“The organisations in this new group are based in more than 15 countries, and most provide critical infrastructure to many more,” the statement said.
The selected organisations represent key sectors including financial services, cybersecurity, communications, energy, water, healthcare and hardware systems.
According to the report, participants in the programme include technology and cybersecurity firms such as Okta, Samsung, SK Hynix and SK Telecom. Financial infrastructure operators including Euroclear, Intercontinental Exchange and Swift have also been included. The rollout further extends to NATO and the European Union Agency for Cybersecurity (ENISA).
Mythos is an advanced cybersecurity AI model developed by Anthropic as part of its frontier AI programme. Designed to analyse large software codebases, the model can identify security vulnerabilities and assist in generating patches to address them. Anthropic describes it as an AI-powered security auditor capable of scanning complex systems and detecting weaknesses that could potentially be exploited by cyber attackers.
The model was first introduced as Claude Mythos Preview in April under tightly controlled conditions through Project Glasswing. Access was initially restricted to around 50 predominantly US-based organisations due to concerns over the model’s powerful code-analysis capabilities and the potential risk of misuse.
Anthropic launched Project Glasswing as a coordinated initiative aimed at using Mythos to identify and patch vulnerabilities in large-scale software systems. The company said early deployments of the model helped uncover more than 10,000 high- or critical-severity security flaws.
The latest expansion follows collaborations with cybersecurity experts, open-source software maintainers and the US government as part of broader efforts to strengthen defensive AI capabilities and improve cyber resilience globally.