India has reportedly joined the growing list of countries evaluating advanced Artificial Intelligence models for cybersecurity preparedness, with the government and major technology companies beginning tests of Anthropic’s cybersecurity-focused AI model, Claude Mythos.
The move comes amid increasing global concerns regarding the capabilities, risks, and potential misuse of highly advanced AI systems in critical digital infrastructure.
According to reports, India’s cybersecurity agencies and leading IT companies are now working together to assess the model’s ability to detect vulnerabilities, identify software flaws, and strengthen cyber defence mechanisms across banking systems, government platforms, and digital public infrastructure.
Anthropic recently introduced its advanced cybersecurity-oriented model under Project Glasswing, initially granting access to a limited group of governments, research organisations, and enterprise partners worldwide. The AI system, known as Claude Mythos, was specifically designed to identify hidden vulnerabilities in software environments and simulate advanced cyber-risk analysis.
As part of the initiative, leading Indian technology firms including Infosys and Tata Consultancy Services are expected to gain access to Mythos Preview for controlled testing and analysis.
The companies will reportedly evaluate their software ecosystems using the AI model to uncover security weaknesses, strengthen protection mechanisms, and improve vulnerability management systems.
One of the major focus areas involves Infosys testing the model’s capabilities on its widely used digital banking platform, Finacle. The AI system is expected to assist in identifying potential security loopholes and generating software fixes for banking infrastructure used by financial institutions globally.
This becomes increasingly important as digital banking systems face growing threats from cybercriminals, ransomware groups, and AI-assisted hacking techniques.
India’s government cybersecurity agency, CERT-In, is also expected to participate in the testing initiative.
According to reports, CERT-In will evaluate several critical digital systems, including:
The objective is to identify vulnerabilities before malicious actors can exploit them.
India’s rapidly expanding digital infrastructure, which supports over 1.4 billion citizens, has made cybersecurity resilience a national priority. Large-scale systems such as Aadhaar, digital payments, e-governance platforms, and banking networks require constant monitoring against evolving cyber threats.
Anthropic’s Claude Mythos has attracted international attention due to its highly advanced vulnerability detection capabilities.
According to Anthropic’s findings, the AI model identified more than 23,000 software vulnerabilities and over 10,000 critical flaws during early testing phases conducted by organisations such as Cloudflare and Mozilla.
These results demonstrated the enormous potential of AI-driven cybersecurity systems, but they also raised concerns regarding misuse if such capabilities were accessed by malicious actors.
Cybersecurity experts globally have warned that advanced AI systems could become double-edged tools. While they can strengthen digital defence systems, they may also enable:
As a result, governments are increasingly prioritizing secure testing frameworks before allowing wider deployment of such AI systems.
Reports suggest that India is currently discussing secure operational access to Claude Mythos with US authorities and partner organizations.
The Indian government has reportedly identified specialised testing facilities and expert teams to evaluate the model’s performance, limitations, and associated cybersecurity risks.
A news agency quoted US Ambassador Sergio Gor saying:
"They (India) have requested access to various things, and as a trusted partner, those are some things that we are definitely looking at."
The discussions are expected to continue as both countries work toward creating a mutually agreed and secure testing environment for the AI model.
The launch of advanced AI models like Claude Mythos has accelerated global discussions around AI governance, cybersecurity policy, and digital resilience.
Several governments and financial regulators worldwide have already advised banks, fintech firms, and enterprise organizations to:
Organizations are being encouraged to perform continuous security testing using AI-powered systems.
Banks and enterprises are strengthening endpoint security, cloud infrastructure protection, and identity verification systems.
Authorities are emphasizing stronger encryption, secure authentication, and stricter access controls.
Earlier in April, Nirmala Sitharaman reportedly held discussions with Indian banks regarding potential risks posed by emerging AI systems, including models like Mythos.
The government highlighted the need for increased vigilance to safeguard:
These developments reflect India’s growing awareness of AI-driven cybersecurity challenges as the country continues its rapid digital transformation.
India has emerged as one of the world’s largest digital economies, supported by initiatives such as:
As more public and private systems become interconnected, cybersecurity risks are becoming increasingly complex.
Experts believe advanced AI systems may eventually become essential tools for:
However, governments and enterprises are also aware that responsible AI governance and oversight will remain critical.
India’s decision to collaborate with leading technology firms and cybersecurity agencies to evaluate Anthropic’s Claude Mythos highlights the growing global urgency around AI-driven cybersecurity preparedness.
As advanced AI systems become capable of identifying thousands of vulnerabilities within critical infrastructure, governments are increasingly treating cybersecurity as a strategic national priority.
The testing initiative involving CERT-In, Infosys, TCS, and other organizations reflects India’s proactive approach toward securing its rapidly expanding digital ecosystem, including Aadhaar, banking systems, and government platforms.
At the same time, the rise of powerful AI cybersecurity tools has intensified global conversations about balancing innovation with responsible oversight and risk management.
As AI technologies continue evolving, India’s early engagement with advanced cybersecurity models like Claude Mythos could help strengthen its digital resilience and prepare critical infrastructure against future cyber threats.