South Korea builds security-focused AI model

South Korea is developing a security-focused AI model intended to give the country sovereign bug-finding capabilities. Deputy Prime Minister and Minister of Science and ICT Bae Kyung-hoon disclosed the project, saying the government wants a domestic system able to rival Anthropic’s Mythos.

The effort follows concerns over access to advanced US-based models. The US government has twice blocked access to Mythos, first by requiring Anthropic to offer it only to American citizens and later by ordering the company to take down its services while Washington investigated allegations of possible dangerous performance problems. Limited access has since been restored for some US allies, but interest in sovereign AI capacity has continued to rise.

South Korea plans to add security-related information to the training corpus for a locally developed frontier model, with a security-capable version expected by the end of 2026. The government has also sought bids for a free chatbot for residents and an agentic application to help people interact with government services, while officials discussed using AI to detect fake news in real time and process public complaints more quickly.