Sinolink: The AI pharmaceutical industry has broad market space; accelerated positioning creates opportunities for wet labs.

Sinolink released a research report stating that leading pharmaceutical companies have already regarded "AI capabilities" as strategic assets equally important as targets and pipelines, and transactions are moving from single-point technology licensing to platform-level M&A. MNCs are deeply embedding AI into the entire chain of R&D, clinical trials, and production in various ways. AI drug discovery can significantly improve early-stage success rates, while late-stage success rates remain to be verified; AI drug discovery is accelerating the expansion of wet lab demand. The AI pharmaceutical industry is in a rapid growth phase, with broad market space. Sinolink's main views are as follows: Traditional drug R&D has long been in a state of "long cycles, high costs, and low success rates" A candidate drug from target confirmation to final approval must go through sequentially advancing stages such as hit discovery, lead optimization, preclinical evaluation, and large-scale clinical trials, with each stage relying on massive resources and repeated validation, resulting in an average R&D cycle exceeding 10 years and cumulative investment per drug exceeding US$2 billion. The intervention of AI drug discovery is expected to significantly compress R&D cycles and expenses: 1) Significantly compress R&D cycles. All R&D stages can benefit, with the drug discovery stage saving the most considerable time, about 7090%; preclinical research can save 5080%; clinical research can save 5060%. 2) Substantially reduce R&D costs. The participation of AI drug discovery in the drug discovery and clinical research stages is expected to save pharmaceutical companies about 50% of costs, corresponding to a scale of about US$54 billion. AI drug discovery has become a battleground, with MNCs accelerating their positioning Global AI drug discovery transactions have continued to heat up since 2024, with upfront payments and total transaction amounts repeatedly hitting new highs. According to statistics from the PharmaCube database, in the first half of 2026, disclosed upfront payments for AI technology transactions globally have reached US$1.8 billion, and total transaction amounts have exceeded US$50 billion, both substantially surpassing full-year 2025 levels. By segment, AI drug discovery has jumped to become the world's third-largest financing track. Facing this track of certain growth, global large pharmaceutical companies (MNCs) have intensively entered since 2024, and in 2026 have pushed their positioning to a new climax: from the perspective of transaction trends. Leading pharmaceutical companies have already regarded "AI capabilities" as strategic assets equally important as targets and pipelines, and transactions are moving from single-point technology licensing to platform-level M&A. MNCs are deeply embedding AI into the entire chain of R&D, clinical trials, and production in various ways. AI drug discovery can significantly improve early-stage success rates, while late-stage success rates remain to be verified Based on existing data, AI-enabled biopharmaceuticals have shown positive signals in Phase I clinical success rates, but it should be noted that the sample size is still small and the conclusion awaits validation with larger-scale data; the advantage in Phase IIIII clinical trials remains to be confirmed. According to IQVIA statistics on the success rates of AI drug discovery pipelines in the two periods of 20202022 and 20232025: AI-enabled drugs have shown success rates above 70% in some Phase I clinical trials (traditional pathways about 50%), with obvious early-stage efficiency gains; however, after entering Phase II, because the sample size is not yet sufficient to form a statistical judgment, the advantage of AI enablement is currently difficult to confirm and still requires the accumulation of more pipeline data for validation. AI drug discovery accelerates the expansion of wet lab demand The DryWet Loop is the most core application paradigm of AI drug discovery at the current stage: through repeated linkage of "AI design prediction (dry lab)" and "automated synthesis testing (wet lab)," the Design-Make-Test-Analyze closed loop is realized, allowing models to continuously self-correct with experimental results. It is currently mainly implemented in early-stage segments such as target validation, molecular design, lead optimization, and ADMET evaluation. The precision of AI models highly depends on high-quality, iterable experimental data for training and correction, and the generated candidate molecules still require experimental synthesis and activity validation before entering the next stage. The same round of experiments must validate more candidates and produce more data, driving the total volume of experimental demand continuously upward. According to Twist's 2026 Investor Day presentation materials, as of March 5, 2026, Twist had cumulatively completed 200,000+ protein expressions and 130,000+ antibody tests in the wet lab stage, and generated more than 7 million data points based on these. Investment recommendations and valuation The AI pharmaceutical industry is in a rapid growth phase, with broad market space. Industry chain companies include: 1) Wet lab service providers: Twist Bioscience, GenScript Biotech, Acrobiosystems, WUXI BIO, etc.; 2) AI drug discovery companies: INSILICO, Jitai Technology, XtalPi; 3) Laboratory services: WuXi AppTec, Pharmaron Beijing, Shanghai Medicilon Inc., etc. Risk warnings Technology and clinical risks; data risks; regulatory risks; commercialization risks; industry and cost risks.