India is among the group of countries whose organisations have gained access to Anthropic’s advanced artificial intelligence cybersecurity model under the company’s expanded Project Glasswing initiative, according to sources cited by news agency PTI.
Sources familiar with the matter were quoted as saying that a handful of Indian entities from both the government and private sectors have been granted access to Claude Mythos Preview, an AI model designed to identify software vulnerabilities and strengthen cyber defences.
The number of Indian organisations currently participating remains limited.
Asked how many Indian entities have secured access to the model, sources told PTI that the figure is “in the single digit for now”.
While the sources confirmed
participation from both government agencies and private firms, they did not disclose the names of the organisations involved.
PROJECT GLASSWING EXPANDS TO MORE COUNTRIES
Project Glasswing has been described by Anthropic as a “collaborative effort to secure the world’s most important software”.
The initiative seeks to bolster cybersecurity by providing select organisations access to Claude Mythos Preview, enabling them to identify weaknesses in software systems before they can be exploited.
According to PTI, when the programme was first announced globally in April, around 50 partner organisations were given access to the model.
Anthropic said those early participants had collectively identified more than 10,000 high- or critical-severity security flaws across various codebases.
The company announced a major expansion of the programme earlier this week.
As part of the latest phase, approximately 150 additional organisations across more than 15 countries have been brought into the initiative.
PTI reported that the new group includes operators of critical infrastructure spanning sectors such as power, water, healthcare, communications and hardware.
In a statement issued on June 2, Anthropic said, “And many of the new partners are vendors – companies or non-profits that maintain codebases that are relied upon by lots of other organisations around the world, including governments.”
The company added that participating organisations are generally those where a successful cyberattack could have significant consequences for large populations.
According to PTI, Anthropic said these are organisations for which a cyber breach could have “far-reaching ramifications”, potentially affecting more than 100 million people in some instances.
The access is expected to help participating entities identify vulnerabilities in digital infrastructure, evaluate cyber defences and speed up efforts to patch vulnerable software systems.
COMPANY CALLS FOR CAUTION ON AI ADVANCEMENT
Alongside the expansion of its cybersecurity initiative, Anthropic has also raised concerns about the rapid pace of artificial intelligence development.
In a recent blog post titled “When AI Builds Itself”, the company argued that policymakers and AI developers should preserve the option of slowing down or temporarily pausing frontier AI development if necessary.
“We believe it would be good for the world to have the option to slow or temporarily pause frontier AI development to enable societal structures and alignment research to keep up with the advances of the technology,” Anthropic said in the blog post, as quoted by PTI.
The company warned that increasingly capable AI systems may eventually acquire the ability to accelerate their own development at a pace that exceeds human oversight.
Separately, ANI reported on June 5 that Anthropic had warned that the evolution of artificial intelligence is approaching a stage where AI systems could potentially design and develop their own successors.
In a report cited by ANI, the San Francisco-based company said it is already “delegating a growing share of AI development to AI systems themselves, which is speeding up our work”.
“Taken far enough, and given enough compute, that trend points to an AI system capable of fully autonomously designing and developing its own successor,” Anthropic said, according to ANI.
