India’s cybersecurity agency, CERT-In, has secured access to Anthropic’s restricted frontier AI model Claude Mythos to test its ability to detect and potentially exploit system vulnerabilities, government sources said, according to a Moneycontrol report. The mod - Instagram
India's cybersecurity agency CERT-In has gained access to Anthropic's restricted frontier AI model Claude Mythos to evaluate its capabilities in detecting and potentially exploiting system vulnerabilities.
According to government sources cited by Moneycontrol, the agency will assess Claude Mythos's ability to identify system weaknesses that could be exploited by malicious actors. This testing represents a significant step in India's approach to leveraging advanced AI technologies for cybersecurity defense.
Testing Capabilities
Claude Mythos, Anthropic's frontier AI model, has been restricted to authorized users only. CERT-In's access marks one of the first instances where a government cybersecurity agency has been granted permission to evaluate such advanced AI systems for defensive purposes.
Related
