MSIT and Anthropic Discuss 'Mythos' Issues: "Collaboration with Global Companies"

MSIT Requests Vulnerability Information Sharing

AISI Discusses Collaboration on AI Model Safety

The Ministry of Science and ICT (MSIT) recently met with Anthropic, a global AI company gaining attention for its high-performance artificial intelligence (AI) model "Mythos," to discuss collaboration measures for addressing security issues and revitalizing the AI industry.

Ryu Jemyung, Second Vice Minister of the Ministry of Science and ICT, is announcing the 'First Phase Evaluation Results of the Independent AI Foundation Model Project' on January 15, 2026, at the Government Complex Seoul in Jongno-gu, Seoul. Photo by Jo Yongjun

On May 11, the Ministry of Science and ICT announced that it had held a roundtable with Anthropic to seek ways to cooperate in the AI and cybersecurity sectors. This meeting was a follow-up to the discussion between Deputy Prime Minister and Minister of Science and ICT Bae and Anthropic CEO Dario Amodei at the "2026 India AI Impact Summit" held in February regarding collaboration in the AI sector. Attendees included Ryoo Jemyung, Second Vice Minister of Science and ICT; Kim Myungjoo, Director of the Artificial Intelligence Security Institute (AISI); Oh Jinyoung, Director at Korea Internet & Security Agency (KISA); and Michael Sellitto, Global Head of Policy at Anthropic.

The roundtable focused on response strategies for cybersecurity issues arising from Anthropic's high-performance AI model, Mythos. The Ministry of Science and ICT proposed that Anthropic actively engage in ongoing practical discussions and cooperation with domestic companies and institutions regarding cybersecurity, and collaborate on the use of AI models in the cybersecurity field. The ministry also requested that Anthropic share information to help South Korea preemptively address cybersecurity vulnerabilities.

Recently, the Ministry of Science and ICT has been exploring response strategies for Mythos, which, due to its outstanding capabilities, could potentially undermine existing security systems. On May 8, Deputy Prime Minister Bae met with domestic AI companies to discuss measures to address cybersecurity concerns related to Mythos. Based on these discussions, the ministry plans to announce countermeasures for AI-related hacking by the end of this month. In addition, the Ministry of Science and ICT is considering participating in "Project Glasswing"—an initiative led by Anthropic to address AI security issues and secure access to Mythos, with AISI at the center of these efforts.

On the same day, the Ministry of Science and ICT and Anthropic also discussed ways to cooperate on domestic AI policies aimed at establishing a foundation of AI safety and trust. During the meeting, the ministry introduced Anthropic to the "Basic Law on AI," which aims to establish an administrative system based on AI and create an AI ecosystem grounded in safety. The ministry emphasized that the Basic Law on AI is attracting global attention as a leading example and requested that Anthropic share its diverse experiences. AISI also discussed potential avenues for collaboration with Anthropic to ensure the safety of generative AI models.

Vice Minister Ryoo stated, "The performance of frontier-level AI models is increasing rapidly, and their applications are expanding. It is important to create an environment where AI can be used safely. We will actively cooperate with global AI leaders to strengthen systems for preventing and responding to AI risks."

This content was produced with the assistance of AI translation services.