Why India Is Seeking Access to Anthropic’s Mythos Despite Security Concerns: What’s at Stake

Days after a high-level review led by Finance Minister Nirmala Sitharaman, the Centre is working on ways to enable Indian companies to access Anthropic's Mythos AI model, according to a report by The Economic Times. The discussions are part of ongoing bilateral talks with the US administration, with a focus on ensuring fair access while safeguarding India’s digital and financial systems.

The government is evaluating both the mechanisms and logistical aspects of such access, while ensuring that critical infrastructure remains protected, the report added.

Rising Cyber Risk Concerns

Speaking at an event, Sitharaman said the Ministry of Electronics and IT is actively engaging with the US government, Anthropic and companies involved in early testing of the model. She flagged that the cybersecurity challenges linked to Mythos could be significant and require close attention.

The governmnet has also asked the Indian Computer Emergency Response Team (CERT-In), the National Critical Information Infrastructure Protection Centre (NCIIPC) and the financial sector to ensure they move fast to protect critical infrastructure such as power grids, telecom networks and banking channels among others that could be vulnerable.

Last week, the Finance Minister asked the Indian Banks’ Association (IBA) to develop a coordinated mechanism for quickly responding to threats linked to the model and directed banks to strengthen their cybersecurity systems by working with specialised agencies and experts.

Notably, the Centre is also considering a broader policy response as more companies develop similar high-capability AI models. It reportedly doesn’t want to favour any particular company by enabling access to some while leaving out others.

It is further considering a policy response to more AI companies launching such models in the future.

Mythos & Risks

Mythos is part of Anthropic's Claude family of AI models, built for complex reasoning and coding tasks. What distinguishes it is its ability to operate across the cybersecurity spectrum, from identifying vulnerabilities to actively exploiting them.

In an April 7 note, Anthropic described the model as a "watershed moment for cybersecurity," stating it had already identified thousands of high-severity vulnerabilities across major operating systems and web browsers, and that it could outperform humans in certain hacking tasks.

Regulators, are concerned that AI systems like Mythos could identify and exploit vulnerabilities faster than institutions can repair them, particularly in banking, where legacy infrastructure remains common, an earlier report by Reuters stated.

Anthropic has said Mythos uncovered vulnerabilities that had persisted for decades.

Officials cited by the Financial Times described the model as a "fundamental change in the playing field," warning it could chain together vulnerabilities at a speed and scale beyond human capability. In finance, Reuters noted, cyber incidents can rapidly spill over into market disruptions and undermine broader confidence in the system.

Mythos AI Trials

Under its Project Glasswing initiative, Anthropic has provided early access to around 40 companies, mostly in the US, to test the model’s capabilities in detecting and fixing security flaws. No Indian company was included in the list.

ET reported earlier that industry body Nasscom wrote to Anthropic seeking inclusion of Indian firms in Project Glasswing. The request highlighted the need for Indian companies to access such tools to strengthen global cybersecurity resilience, especially as their software supported systems worldwide.

The companies that tested Mythos said it can identify tens of thousands of vulnerabilities, compared with around 500 detected by Anthropic’s earlier model, Opus 4.6, marking a nearly 20-fold jump in a single generation.