According to a research report from China Securities Co., Ltd. (ASX: 601066), the AI pharmaceutical sector is experiencing rapid iteration and transformation. Algorithm updates and computing power support have laid a solid technical foundation for AI applications in drug development. Traditional drug development processes are time-consuming, costly, and have low success rates. As new drugs targeting existing targets are approved successively, the difficulty of new drug development is gradually increasing. Currently, AI has demonstrated efficiency value in early stages such as structure prediction, target screening, virtual screening, molecule generation, and multi-parameter optimization, and is expected to narrow the range of candidate molecules, reduce low-value experiments, and lower the discovery cost per qualified Hit. Increased investment in related dry lab activities and sustained growth in wet lab demand are expected to continuously bring incremental opportunities to the downstream supply chain. The main viewpoints of China Securities Co., Ltd. are as follows:
AIDD First Reshapes Drug Discovery and Preclinical R&D Processes
Small molecule discovery is evolving toward an AI-participating DMTA (Design-Make-Test-Analyze) closed loop, with notable benefits in virtual screening, molecule generation, multi-parameter optimization, ADMET (absorption, distribution, metabolism, excretion, toxicity) pharmacokinetic and safety property prediction, and synthetic route design. Large molecule R&D is gradually forming a DBTL closed loop starting from sequence and structure design, further encompassing function prediction, antibody generation, and multi-parameter optimization. High-quality experimental data and high-throughput validation capabilities are becoming key constraints for further AI pharmaceutical development, and supply chain increments are expected to be realized first in early experimental stages. The model's dependence on high-quality experimental data is further strengthened, requiring continuous gene synthesis, library construction, protein/antibody expression, binding and function testing, and multi-omics analysis, with experimental results fed back into model training. Therefore, as AI model iteration accelerates, the number of candidates per design increases, and more pharmaceutical companies embed AI into early R&D processes, wet lab demand is expected to continue growing. Supply chain segments such as gene synthesis, protein expression, research reagents, and CROs are expected to continuously benefit from incremental markets brought by increasing AI pharmaceutical penetration. Service providers with high-throughput, automated, standardized experimental and structured data delivery capabilities are more likely to capture these increments.
The Industry Is Shifting from Model Capability Validation to Clinical and Commercial Value Validation, with Positive Signals from Leading Projects and Collaborations, but Overall Still in Early Stages
Representative pipelines such as Rentosertib and GB-0895 have entered mid-to-late clinical stages, providing higher-stage samples for observing the clinical translation capability of AI-assisted drug discovery. MNC external collaborations, asset licensing, and platform deployments continue to increase. AI pharmaceutical companies mainly adopt three models: platform and software licensing, AI R&D services, and self-developed pipelines and asset licensing. These three models differ significantly in cash flow formation timing, clinical risk, and potential value space.
Related Companies
The AI pharmaceutical industry has broad long-term growth space, and related supply chain companies include: 1) Wet lab service providers: GenScript Biotech, Twist Bioscience, WuXi Biologics, Sino Biological, ACROBiosystems, and Novoprotein; 2) Overseas giants: Anthropic, Isomorphic Labs, and Chai Discovery; 3) AI pharmaceutical targets: Insilico Medicine, JITAI Technology, and XtalPi.
Risk Warnings
Risks include investment and financing falling short of expectations, industry policy risks, R&D falling short of expectations, approval falling short of expectations, sales falling short of expectations, and intensifying market competition.
Comments