AI-Driven Drug Development Faces a Real Test... The Key Lies in Experimental Data [Weekend Money]

Global Big Pharma Expands AI Partnerships with Big Tech
Faster Progress Does Not Guarantee Clinical Success
Proprietary Data Becomes the Core Asset

The competition for leadership in new drug development leveraging artificial intelligence (AI) is shifting from analysis algorithms to securing proprietary experimental data.

Recently, Jisoo Lee, a researcher at Daol Investment & Securities, commented, "Investment is now focused on systems that connect AI analysis with actual experiments," and added, "In the AI era, it's not enough to analyze existing data; the core competitive edge lies in the ability to directly generate and utilize novel experimental data."

Big Pharma Joins Hands with AI Giants... Building Positive-Feedback Data Systems

Global pharmaceutical giants see AI technology as a key to streamlining research and development (R&D) and are expanding partnerships with big tech companies across the board. Novo Nordisk has decided to pilot Anthropic's "Claude Science" in some R&D tasks, and Bristol Myers Squibb (BMS) has adopted "Claude Enterprise" as a common AI platform throughout its global operations. Merck & Co., Inc. (MSD) has signed a contract with Google Cloud worth up to 1 billion dollars to leverage accumulated data.

Beyond simple data analysis, a virtuous cycle that combines real-time experiments and AI is also being actively established. Researcher Lee explained, "Eli Lilly and NVIDIA have announced a joint investment of up to 1 billion dollars over five years to build a system connecting AI analysis and laboratory experiments," and continued, "This system allows AI to learn from experimental results and design subsequent experiments with greater sophistication."

However, Lee noted, "Even if the pace of identifying candidate molecules through AI is accelerated, this does not necessarily translate into higher clinical success rates," stressing that "the safety and efficacy in humans must be verified independently." Ultimately, companies that successfully integrate hands-on research experience and validated experimental capabilities with AI are expected to deliver tangible results.

Korean Companies Strike Back... Expanding Proprietary Platforms and Data Pools

This shift is expected to create new opportunities for domestic bio companies that have built technological expertise in specific diseases and drug fields. Systematizing proprietary data and expanding data pools through joint research with universities and hospitals can provide a unique AI competitive advantage. Even failed experimental data from the past are expected to be reborn as core assets that reduce trial and error in future studies.

In fact, Hanmi Pharmaceutical combined peptide design expertise and experimental data on activity and selectivity of candidate substances with its proprietary "HARP-PSAR" platform to identify and optimize the obesity candidate "HM17321."

Examples of combining external platforms are also increasing. YBiologics is developing immuno-oncology drugs by integrating the AI drug design technology of Galux, an AI-based drug discovery company, with its own antibody platform. Protina has established a system that feeds antigen interaction data into AI by linking its AI antibody design platform (AbGPT-3D) with large-scale experimental validation (SPID), and has signed a joint technology development and technology transfer option agreement with Samsung Bioepis.

Researcher Lee remarked, "Companies equipped with proprietary data and experimental capabilities can broaden their collaboration opportunities as research partners for AI companies," adding, "Such data competitiveness is ultimately expected to lead to business outcomes, including global partnerships and technology transfers."

This content was produced with the assistance of AI translation services.