CICC: AI4S reshapes the intelligent paradigm of new drug R&D; bullish on differentiated platforms and pipeline delivery capability

CICC released a research report stating that AI for Science (AI4S, AI-driven scientific research) continues to gain momentum, with biomedicine being a core application scenario. It recommends continued attention to three types of companies: 1) companies that rely on differentiated technology platforms to effectively improve R&D efficiency; 2) companies capable of continuously producing high-quality pipelines, expected to achieve key clinical progress, and possessing out-licensing potential; 3) companies with experimental feedback capabilities and automated experimentation systems that can accommodate incremental validation demand. CICC's main views are as follows: The efficiency-enhancing capability of AI drug discovery is gradually emerging, with platform capabilities and drug asset value being validated in tandem AI has progressively penetrated areas such as protein structure prediction, virtual screening, molecular design, and lead optimization, potentially reducing experimental trial-and-error and shortening R&D cycles; leading pipelines are entering mid-to-late-stage clinical development, providing more basis for evaluating their R&D value. The bank believes that platforms with differentiated technologies that can continuously improve R&D efficiency deserve attention, and key clinical progress and potential licensing opportunities are also expected to further reflect asset value. Capital and industry investment continue to increase, and platform competition increasingly depends on data and experimental capabilities Tech giants are expanding from computing power and biological foundation models to scientific research agents, while MNCs are accelerating the integration of internal data, experimental systems, and ongoing projects with AI, driving the transition from technological exploration to R&D application. However, insufficient high-quality causal data, difficulty in validating target mechanisms, and lengthy synthesis and testing cycles still constrain the conversion of computational advantages into R&D outcomes. The bank believes that the key to establishing differentiated advantages for platforms lies in continuously generating proprietary experimental data, accelerating validation feedback, and using data accumulation to support subsequent model optimization. AI expands the scope of R&D exploration, and the industry chain is expected to benefit simultaneously from validation demand growth and service capability upgrades Lower design costs can enable more projects to advance and generate experimental demand; CROs can also apply AI in combination with existing experimental data, screening, and structural biology technologies to expand service depth and improve delivery efficiency. The bank is bullish on companies that can leverage synthesis, expression, and functional validation capabilities to accommodate incremental demand, as well as companies that utilize differentiated platforms and data accumulation to continuously improve R&D efficiency and service depth. Risk reminders New drug R&D risks; intensifying market competition; business fluctuations or commercial returns below expectations; constraints on data usage and computing power conditions.