수십만 물질서 ‘탈모 효능’ 소재 콕…AI, 22개월 걸릴 일 하루만에

The search for a single substance capable of alleviating hair loss once required an extensive 22-month period among the 42,000 available chemical compounds. However, the integration of artificial intelligence (AI) into the process has significantly shortened this timeline. AI not only identified the promising substance, 'Lamdisil,' within a single day but also autonomously designed the subsequent experiments based on initial experimental outcomes.

At the 'AI Talk Concert 2026' held on the 14th, LG AI Research Institute showcased the successful use of AI to discover 'Lamdisil,' a substance with potential hair loss alleviating effects from a vast pool of candidates. Traditional methods required researchers to compare each candidate material one by one, but AI can tirelessly analyze multiple substances 24/7, identifying potential candidates much faster and significantly reducing the time spent on the exploration phase.

LG AI Research Institute plans to employ AI 'autonomous experiment stations,' which can independently conduct experiments and learn from them, further accelerating progress in scientific and medical fields. This institute is working with the Vanderbilt Medical Center in the United States on an AI-based 'cancer agent' that can predict a patient's genetic activity status and suggest personalized treatment strategies in less than a minute.

This development extends to a broader range of cancers, including colon and lung cancers. The AI's role doesn't end with automation. AI analyzes experimental results, determines the next experiment, and executes it through robots in a 'loop in the lab' autonomous experiment system. This continuous cycle of experiment → analysis → next experiment allows research to proceed around the clock.

Other South Korean companies like Samsung Bio's HARP platform and CJ Bio are also leveraging AI in their drug development processes. HARP is used to design anti-obesity drug candidates, and a technology transfer agreement worth up to $235 million with American Genentech was recently signed. CJ Bio is also exploring a rare neurodegenerative disease drug candidate in pre-clinical stages.

In the US and Hong Kong, AI-driven pharmaceutical company Insilico Medicine entered clinical stage 3 for their drug candidate 'Rentosip' targeting pulmonary fibrosis after AI learned from patients' medical records, protein information, and medical literature. Another AI is creating molecular structures capable of inhibiting specific target proteins.

American pharmaceutical company Reckitt is conducting over 2 million cell experiments annually using AI. This automated approach to cell experiments and image analysis significantly expands the scale of research. The integration of AI and automated labs for large-scale cell experiments and image analysis is paving the way for a new era in drug development.

The Korean government is also investing heavily in AI and bio-integrated research and development. The Ministry of Science, ICT & Future Planning launched the 'K-Complete' project, aiming to tackle national scientific and technological challenges using AI by 2035. As part of this initiative, six advanced bio-autonomous experiment stations will be established, with a total investment of 495 billion won over three years.

The 'K-Cell Universal Autonomous Experiment Platform,' focusing on various fields, has been selected as one of the key projects.