What is the story about?
As cyberattacks become more sophisticated and critical infrastructure increasingly depends on software, the race to identify security flaws before hackers do has become a national priority. India, which manages a diverse digital ecosystems spanning banking, telecom, healthcare and public services, has long argued that access to advanced cybersecurity tools should not be limited to a handful of countries.
Now, that effort appears to have paid off.
Anthropic has expanded access to its powerful AI cybersecurity model, Mythos, bringing India into a select group of nations that can use the technology to detect critical software vulnerabilities. The move comes months after Indian authorities reportedly engaged with both Anthropic and the United States to secure what they described as equitable access to the model.
For India, the development is about far more than gaining access to another AI tool. It represents an opportunity to strengthen cyber resilience across sectors that millions rely on every day, while ensuring the country is not left behind as artificial intelligence reshapes global cybersecurity.
The announcement is part of Anthropic's broader expansion of Project Glasswing, a joint industry initiative aimed at using AI to identify and fix critical software vulnerabilities before they can be exploited by malicious actors.
At the centre of the programme is Mythos, which Anthropic describes as its most capable cybersecurity model to date. Earlier this year, the company granted a small group of around 50 partners, including US government agencies and select organisations, access to a preview version of the model.
The latest expansion adds approximately 150 organisations across more than 15 countries. The new cohort includes entities operating in sectors such as energy, water, healthcare, telecommunications and hardware manufacturing, industries where a successful cyberattack could have far-reaching consequences.
According to reports, the expanded access programme includes organisations and institutions from countries including India, Japan, South Korea, Germany, France, Australia, Canada and New Zealand. Several major global organisations, including cybersecurity agencies and technology companies, have also reportedly been granted access.
Anthropic says many of the participating organisations maintain software and digital systems that underpin services relied upon by governments, businesses and citizens worldwide. In many cases, a major compromise could affect hundreds of millions of people.
For India, access to Mythos could significantly enhance efforts to secure critical digital infrastructure.
The country has undergone a rapid digital transformation over the past decade, with banking, digital payments, telecommunications, healthcare and government services becoming increasingly interconnected. While this has delivered convenience and economic benefits, it has also expanded the potential attack surface available to cybercriminals.
Mythos is designed to identify vulnerabilities on a scale far beyond traditional security tools. According to Anthropic, the model can uncover thousands of previously unknown weaknesses across operating systems, applications and web platforms. Internal testing reportedly showed it discovering tens of thousands of vulnerabilities, dramatically exceeding the capabilities of earlier systems.
Some of the flaws identified by the model are said to have existed for decades without being detected, highlighting the difficulty of securing complex software ecosystems.
For Indian organisations responsible for critical infrastructure, the technology could help uncover hidden risks before attackers exploit them. This is particularly important in sectors such as banking, power distribution and telecommunications, where disruptions can have widespread economic consequences.
However, the technology also raises concerns. Security researchers and regulators have warned that advanced AI models could become double-edged swords. While they can help defenders identify weaknesses, they could also enable attackers to discover and chain together vulnerabilities at unprecedented speed.
That concern is especially relevant for countries with large legacy technology environments, where outdated systems remain in operation.
Still, for India, gaining access to Mythos ensures it has a seat at the table as AI begins to redefine cybersecurity. Rather than watching from the sidelines, the country can now participate in shaping how these powerful tools are used to protect critical systems in an increasingly digital world.
Now, that effort appears to have paid off.
Anthropic has expanded access to its powerful AI cybersecurity model, Mythos, bringing India into a select group of nations that can use the technology to detect critical software vulnerabilities. The move comes months after Indian authorities reportedly engaged with both Anthropic and the United States to secure what they described as equitable access to the model.
For India, the development is about far more than gaining access to another AI tool. It represents an opportunity to strengthen cyber resilience across sectors that millions rely on every day, while ensuring the country is not left behind as artificial intelligence reshapes global cybersecurity.
Anthropic expands Mythos access to India and other nations
The announcement is part of Anthropic's broader expansion of Project Glasswing, a joint industry initiative aimed at using AI to identify and fix critical software vulnerabilities before they can be exploited by malicious actors.
At the centre of the programme is Mythos, which Anthropic describes as its most capable cybersecurity model to date. Earlier this year, the company granted a small group of around 50 partners, including US government agencies and select organisations, access to a preview version of the model.
The latest expansion adds approximately 150 organisations across more than 15 countries. The new cohort includes entities operating in sectors such as energy, water, healthcare, telecommunications and hardware manufacturing, industries where a successful cyberattack could have far-reaching consequences.
According to reports, the expanded access programme includes organisations and institutions from countries including India, Japan, South Korea, Germany, France, Australia, Canada and New Zealand. Several major global organisations, including cybersecurity agencies and technology companies, have also reportedly been granted access.
Anthropic says many of the participating organisations maintain software and digital systems that underpin services relied upon by governments, businesses and citizens worldwide. In many cases, a major compromise could affect hundreds of millions of people.
What does it mean for India?
For India, access to Mythos could significantly enhance efforts to secure critical digital infrastructure.
The country has undergone a rapid digital transformation over the past decade, with banking, digital payments, telecommunications, healthcare and government services becoming increasingly interconnected. While this has delivered convenience and economic benefits, it has also expanded the potential attack surface available to cybercriminals.
Mythos is designed to identify vulnerabilities on a scale far beyond traditional security tools. According to Anthropic, the model can uncover thousands of previously unknown weaknesses across operating systems, applications and web platforms. Internal testing reportedly showed it discovering tens of thousands of vulnerabilities, dramatically exceeding the capabilities of earlier systems.
Some of the flaws identified by the model are said to have existed for decades without being detected, highlighting the difficulty of securing complex software ecosystems.
For Indian organisations responsible for critical infrastructure, the technology could help uncover hidden risks before attackers exploit them. This is particularly important in sectors such as banking, power distribution and telecommunications, where disruptions can have widespread economic consequences.
However, the technology also raises concerns. Security researchers and regulators have warned that advanced AI models could become double-edged swords. While they can help defenders identify weaknesses, they could also enable attackers to discover and chain together vulnerabilities at unprecedented speed.
That concern is especially relevant for countries with large legacy technology environments, where outdated systems remain in operation.
Still, for India, gaining access to Mythos ensures it has a seat at the table as AI begins to redefine cybersecurity. Rather than watching from the sidelines, the country can now participate in shaping how these powerful tools are used to protect critical systems in an increasingly digital world.
