South Korea building sovereign cybersecurity AI after US Mythos export controls (en.yna.co.kr)

South Korea is set to develop its own sovereign artificial intelligence (AI) model focused on cybersecurity by the end of this year, as announced by Science Minister Bae Kyung-hoon. This initiative is a direct response to rising digital security threats and recent U.S. export controls on advanced AI models like Anthropic's Mythos 5, which specializes in detecting vulnerabilities. The South Korean government aims to enhance its current AI capabilities by training its sovereign model on security-specific data to better address these evolving threats and solidify the nation’s position in the AI competitive landscape. The significance of this development extends beyond immediate cybersecurity needs; it highlights South Korea's ambition to rank higher in AI competitiveness globally. Minister Bae emphasized the necessity of establishing a frontier model equivalent to Mythos in the long term. Additionally, discussions during the policy briefing included the need for legislation to support ethical hacking practices, or "white hacking," to proactively identify vulnerabilities in systems. This strategic direction not only strengthens South Korea's cybersecurity framework but also fosters an environment of innovation in AI, targeting both defensive measures and expanded AI capabilities within the state.