India gets access to Anthropic’s restricted Mythos Claude under Project Glasswing

India will soon gain access to Claude Mythos — an artificial intelligence model developed by Anthropic and deemed ‘too powerful’ for public release. The move comes less than a day after the AI startup filed for an initial public offering at nearly $1 trillion valuation. Mythos had sparked global concern following its launch in April, and financial institutions in India remain wary about its powerful cybersecurity skills.

According to a Reuters report, Anthropic has confirmed a wider rollout of its latest AI model under Project Glasswing. The San Francisco-based ⁠startup said its partnership would expand to include around 200 organisations — with each member required to meet certain security requirements before accessing Claude Mythos Preview.

Details remain scarce but Reuters quoted an Anthropic spokesperson to confirm that the expansion includes government organisations. The new Glasswing partners are based in more than 15 countries and boost defenses in the healthcare, power, water, communications and hardware industries. Anthropic estimated that a big attack on partner organizations could affect more than 100 million people.

India gains access to Mythos

A separate report by Financial Times revealed that India was among the countries getting access to the Mythos AI model. The list reportedly includes nations within the “Five Eyes” intelligence alliance (such as Canada, Australia and New Zealand) as well as multiple European and Asian countries. Financial Times quoted a source to report that France, Germany, Italy, Switzerland, the Netherlands, Spain, Belgium, Sweden, Japan and South Korea were being granted access alongside India.

Companies that can access Mythos as part of the expansion reportedly include US tech group Okta and South Korean companies Samsung, SK Hynix and SK Telecom.

The South Korean Science Ministry also confirmed access on Wednesday morning along with major Korean companies. The Ministry of Science and ICT said in a statement it had been working continuously with Anthropic ⁠and confirmed KISA’s participation in the initiative, which is ‌aimed at using frontier AI models to identify and help fix cybersecurity vulnerabilities.