当英伟达（NVIDIA）创始人兼CEO黄仁勋在J.P. Morgan医疗健康大会和GTC大会的聚光灯下反复提及这一论断时，生物学正在从一门依赖试错的 实验科学，转变为一门可预测、可编程的数据科学。 而英矽智能（Insilico），正是这个被黄仁勋多次点名的"第一性原理"践行者。 "未来，所有的生物学都将以'计算机模拟'（in silico）为起点，并以'计算机模拟'为很大程度的终点。"（Almost everything will largely start in silico, largely end in silico） 2025年12月30日，英矽智能（3696.HK）在香港联合交易所主板正式挂牌，开盘价报35港元，较发行价升45%，市值达到195亿港元。 | ■ 分时 日 周 月 季 年 1分 5分 15分 30分 60分 多 周 周 安 安 | 竟 九转 则自选 ▼ | | 英矽智能-B 03696 | | 最新 | 35.000 均价 | 38.347 | | --- | --- | --- | --- | --- | --- | --- | --- | | 年67回服-B 分析 15H1: 38 ...

Core Insights - The article posits that while we are in an era of unprecedented technological prosperity, innovation is becoming increasingly difficult to achieve, with AI potentially serving as the key to overcoming this bottleneck [1][8]. Group 1: Innovation Challenges - The cost and difficulty of innovation have escalated globally, affecting various industries [3][5]. - R&D spending in the chip industry is projected to be 18 times higher than in the 1970s by 2024, while the pharmaceutical industry has seen an 80-fold decrease in the number of new drugs developed per $1 billion invested over decades [4][5]. - The overall productivity of R&D in U.S. companies has been declining since the 1950s, a trend observed globally [5][8]. Group 2: AI as a New Pathway - AI is positioned as a transformative force that can propose "questions humans would not think of" and "paths humans would not choose" in the innovation process [11][17]. - AI's ability to generate numerous design candidates and explore unconsidered paths is highlighted, with examples from various fields such as protein synthesis and retail space design [15][16]. Group 3: Revolutionizing Validation - The validation phase of R&D, often the most time-consuming, can be expedited through AI, which can simulate and predict outcomes much faster than traditional methods [19][24]. - AI models, known as surrogate models or digital twins, can replicate complex physical processes with minimal computational resources, significantly reducing the time and cost of validation [26][30]. Group 4: AI's Role in Knowledge Integration - AI is redefining the management of implicit knowledge within organizations, enabling the aggregation of insights from various sources, including social media and internal communications [40][41]. - The ability of AI to process vast amounts of data allows for the identification of trends and user needs that may not be immediately apparent to human researchers [42][44]. Group 5: Industry-Specific Applications - In software and gaming, AI is automating code generation and content creation, significantly reducing development time [54][55]. - In life sciences, AI is being utilized to identify molecular targets and predict protein structures, enhancing drug discovery processes [57][60]. - In materials science, AI accelerates the discovery of new materials by predicting properties without physical experiments [62][63]. - In aerospace and complex manufacturing, AI integrates multi-disciplinary engineering processes, improving design accuracy and efficiency [66][67]. - In consumer goods, AI analyzes consumer feedback to inform product development, reducing the risk of market failure [70][71]. Group 6: Future of Innovation - The article concludes that AI is not just a tool but a collaborative partner in the innovation process, transforming R&D into a co-creative ecosystem rather than a linear workflow [74][80]. - The potential for AI to reverse the decline in innovation rates could significantly impact economic growth and societal well-being in the future [81][82].