The government is moving to build an AI drug discovery research system that links AI models with automated laboratories to identify and validate new drug candidates.

The Ministry of Science and ICT said on Tuesday it held a launch ceremony in Daegu for the pilot task force to build an AIxBio innovation research hub for synthetic drugs and has started operating an AI-based drug discovery innovation hub.

The core is a Lab-in-the-Loop approach that connects AI models with real experimental environments.

AI models propose drug candidates and experimental conditions based on bio data, and an automated experimental platform validates them. The experimental results are fed back into AI model training to improve performance.

The plan is to build a circular research system in which AI models and experimental systems repeatedly interact, accumulating research data and improving performance, going beyond the use of AI as a research support tool.

The first pilot hub will be built in Daegu, targeting synthetic drugs.

Kyungpook National University will oversee the project, with Kyungpook National University Hospital, the Daegu-Gyeongbuk Medical Innovation Foundation and Univab participating.

Kyungpook National University will be responsible for building computing infrastructure for AI model development and fostering AI-bio convergence talent. Kyungpook National University Hospital will build a system for using medical data based on patient data covering about 2,000,000 people.

The Daegu-Gyeongbuk Medical Innovation Foundation will work with HLB Life Science, Arontier, Molcube and Ablelabs to set up automated experiments and high-speed validation infrastructure. Univab will be responsible for developing a foundation model to be used in drug discovery.

The Ministry of Science and ICT plans to expand the innovation hub to other AI-bio fields based on the results of the pilot project.

Oh Dae-hyun (오대현), director general for Future Strategic Technology Policy at the Ministry of Science and ICT, said, "Future competitiveness in drug discovery depends on how effectively AI technology and data are connected to and used in real experimental environments." He said, "Next year, we will expand into additional AI-bio fields based on the results of the pilot hub."