专访德勤孙晓臻：抢占“AI+健康”制高点，寻找差异化生死时速

Core Insights - The AI application in healthcare and pharmaceuticals is rapidly transitioning from a "technical concept" to "scaled implementation," with the Chinese AI healthcare market expected to exceed 20 billion yuan by 2025 and 100 billion yuan by 2030, reflecting a compound annual growth rate (CAGR) of 43.2% [1][2] - The global AI pharmaceutical market is projected to surpass 50 billion USD, with significant growth anticipated in drug discovery and medical imaging, which together account for over 50% of the market [3][4] - AI is seen as a growth engine in the healthcare and pharmaceutical sectors, with innovative models emerging under scenario-based support, although commercialization still requires validation [4][5] Market Trends - The AI+ healthcare market is experiencing rapid growth, with an expected annual compound growth rate exceeding 29%, and the market size projected to reach 70 billion USD by 2032 [3][4] - Major pharmaceutical companies are shifting focus from merely following trends to seeking differentiated new targets or optimizing existing ones to establish genuine technological barriers in international markets [1][2] Investment Landscape - Despite macroeconomic adjustments, leading companies in the AI+ healthcare sector continue to attract investment, with significant funding rounds reported, such as Insilico Medicine's 123 million USD Series E financing [4][5] - The investment logic is transitioning from early speculation to a focus on platform capabilities and commercial viability [4][5] AI's Role in Drug Discovery - AI significantly enhances efficiency in target discovery by integrating multi-omics, literature, and databases, thereby shortening the "concept to validation" cycle [2][6] - AI can systematically evaluate the druggability of targets, optimizing resource allocation at earlier stages [2][6] Clinical Trial Dynamics - AI has improved the success rate of Phase I clinical trials from 40%-65% to 80%-90%, but regulatory requirements pose challenges for dynamic learning algorithms [6][7] - The trend is shifting towards a "small and fast" pipeline strategy, emphasizing rapid validation and decision-making, which is being adopted by both AI startups and traditional pharmaceutical companies [7][8] Evaluation of AI Platforms - The focus of pharmaceutical companies is shifting from the quantity of pipelines to the learning capabilities and sustainable output of AI platforms [8][9] - Key metrics for evaluating AI platforms include the breadth and quality of training datasets, cross-target generalization ability, and the actual conversion rates of generated molecules [8][9] Future Outlook - The integration of AI in healthcare is expected to lead to more collaborative models between traditional pharmaceutical companies and AI startups, blurring the lines between the two [7][9] - The ability to create a closed-loop validation system, including automated experimental platforms, is becoming a critical factor in assessing the long-term value of AI platforms [9]