India seeks access to Anthropic's Mythos AI model as government weighs cybersecurity risks

India is negotiating with the US and Anthropic for access to Mythos, an AI that can detect tens of thousands of software vulnerabilities. No Indian firms were included in the initial 40-company access program.

Categorized in: AI News Government
Published on: Apr 29, 2026
India seeks access to Anthropic's Mythos AI model as government weighs cybersecurity risks

India pushes for access to Anthropic's Mythos AI amid security concerns

The Indian government is negotiating with the US and Anthropic to secure access to Mythos, a new AI system that can identify tens of thousands of software vulnerabilities. Officials say discussions focus on equitable access for domestic companies while protecting critical infrastructure.

Finance Minister Nirmala Sitharaman flagged the cybersecurity risks at a recent industry event. She said the Ministry of Electronics and Information Technology is actively engaged with US authorities, Anthropic, and vendors testing the model.

Sitharaman recently convened a meeting with banks, cyber agencies, and IT Minister Ashwini Vaishnaw to assess the situation. The government has asked key agencies-including the Indian Computer Emergency Response Team and the National Critical Information Infrastructure Protection Centre-to strengthen safeguards around power grids, telecom networks, and banking systems.

What makes Mythos different

Mythos can detect tens of thousands of zero-day vulnerabilities, far exceeding the roughly 500 identified by Anthropic's earlier Opus 4.6 model. Early feedback suggests the system can identify and exploit these weaknesses at scale.

Access has been tightly controlled through a programme called Project Glasswing, which granted early access to around 40 companies, mostly based in the United States. No Indian firms were included in that initial group.

The dual challenge

India faces two competing pressures. Domestic companies risk falling behind in accessing advanced generative AI and LLM tools. At the same time, the government must build defences against the threats such systems could introduce.

Officials are also considering a broader policy framework for handling future AI releases. One concern is that companies could roll out similarly powerful systems without advance notice, leaving governments with limited time to respond.

The issue reflects broader questions about AI for government agencies-how to access cutting-edge capabilities while managing the risks they create.


Get Daily AI News

Your membership also unlocks:

700+ AI Courses
700+ Certifications
Personalized AI Learning Plan
6500+ AI Tools (no Ads)
Daily AI News by job industry (no Ads)