LG Corp. (003550.KS) has unveiled artificial intelligence technology that shortens the design of personalized cancer treatment strategies from more than four weeks to just one day. The development suggests that the convergence of AI and biotechnology — the core future businesses identified by LG Chairman Kwang-mo Koo — is beginning to generate meaningful synergy.
LG AI Research announced Monday that it presented the results of its "Cancer Agentic AI" research, jointly developed with Vanderbilt University Medical Center in the United States, at the American Association for Cancer Research (AACR 2026) annual meeting held in San Diego from April 17 to 22 local time. The technology performs the entire process from cancer patient tissue analysis to treatment strategy design within a single day.
At the core of the system is the pathology AI EXAONE Path. Using just a single tissue pathology image, the technology predicts cancer gene activity within one minute at world-leading accuracy, reducing unnecessary tests and identifying early on which patient groups are eligible for targeted drug therapy. Multiple AI agents based on EXAONE are designed to collaborate organically with medical professionals. Moving beyond single question-and-answer exchanges, each AI sequentially performs tasks such as tissue analysis, cancer gene activity verification, drug response validation, and treatment strategy design, then hands off to the next agent.
"The AI agents have created a brain that will innovate personalized cancer treatment through collaboration with medical professionals," said Jang Jong-sung, head of LG AI Research's Bio Intelligence Lab. Professor Hwang Tae-hyun of Vanderbilt University emphasized, "A collaborative model in which AI processes vast amounts of data and medical professionals make the final decisions will deliver greater outcomes in clinical settings."
The system is designed so that diagnoses and recommendations become more refined based on accumulated data as patient cases grow. The two organizations plan to expand the application of agentic AI to various cancer types, starting with gastric cancer and extending to colorectal and lung cancers. Building on this conference, they plan to continue substantive partnership discussions aimed at delivering personalized treatments and reducing the cost and duration of clinical trials.
