"High-quality data is essential for successful drug development," Jeong Yoon-taek, head of the Pharmaceutical Industry Strategy Research Institute, stated on May 14 at the 16th Global Healthcare Forum (2026 GHF) held at the Korea Press Center in Seoul. He emphasized the need for a transition to an AI-driven pharmaceutical industry.
During his presentation, Jeong noted that the pharmaceutical and biotech sectors are entering the 'Pharma 5.0' era. He explained that the industry is facing challenges such as declining productivity in drug development and the expansion of next-generation modalities, making it difficult to maintain competitiveness with traditional development methods.
"Without vast amounts of high-quality data, AI models cannot learn or predict effectively," he said, adding that the principle of 'capturing data to capture new drugs' is a key challenge in the era of AI drug development.
Jeong highlighted the rapid changes in the drug development paradigm. In the past, candidate substances were identified primarily through researchers' experience and repetitive experiments. Now, the industry is shifting towards using extensive big data, including genomic, clinical, and literature data, with AI playing a crucial role in identifying candidate substances and optimizing clinical strategies.
He also discussed the evolution of AI technology, which has progressed from rule-based systems to deep learning and generative AI, and is now advancing into autonomous AI and quantum technology integration.
Previously, drug candidates were designed through computer-based virtual experiments, but now, laboratories are expanding to include robotic automation for actual substance synthesis and validation.
Jeong stated that the increased use of AI is improving the efficiency of drug development. The introduction of AI-based predictive models and automation technologies is expected to shorten clinical trial durations, reduce development costs, and accelerate commercialization.
"AI is not just speeding up research; it is transforming the entire drug development value chain," he said, noting that the use of AI is expanding from candidate discovery to preclinical, clinical, production, and distribution stages.
He also pointed out the strategic shifts among global pharmaceutical giants. These companies are accelerating collaborations and mergers with AI-based platform firms, competing to build open ecosystems centered around data and platforms. He observed a shift from traditional vertically integrated structures to expertise-based network models in the industry.
In contrast, Jeong criticized the limitations of the domestic AI drug development ecosystem. He noted that investments in South Korea are still heavily focused on early-stage candidate discovery, exacerbating the 'death valley' issue where foundational AI technologies fail to transition into medical practice. Many companies are struggling at the commercialization stage due to a lack of organic connections between funding, data, and validation systems.
Jeong emphasized that the key to competing in AI drug development is ultimately data competitiveness. He stressed the importance of alleviating data monopolies among companies and establishing collaborative utilization systems.
He cited the 'MELODI' case and collaborative learning models, stating, "We need to cultivate specialized talent, introduce standardized contract systems, and foster global partnerships. The government must also establish continuous support systems."
* This article has been translated by AI.
Copyright ⓒ Aju Press All rights reserved.