GMAsia
    🇨🇳China·AI News·9 Sept 2026·via SCMP

    As China embraces AI, only 24% of US pharma execs bet on it for drug-making

    A new Citi survey indicates that while pharmaceutical companies are widely adopting artificial intelligence in drug research and development, executives have tempered expectations for its impact on drug success rates. The study found that 72 percent of companies are either scaling or have fully scaled AI across R&D. However, only 24 percent of respondents anticipate a significant improvement in drug success probability due to AI. The survey also noted that 34 percent of executives believe human biology and clinical judgment will remain bottlenecks, regardless of AI deployment. This suggests a more cautious outlook on AI's transformative potential in drug making than initially perceived.

    Nexa's Summary

    The real story here is not that AI is failing in drug discovery, but that the industry is realistic about its limitations. While 72 percent of companies are scaling AI in R&D, only 24 percent of US pharma executives expect a significant improvement in drug success probability. This suggests that while AI can accelerate discovery, human expertise in clinical judgment remains critical. For Asian markets, this points to a need for continued investment in both AI capabilities and skilled human talent in biotech and pharma. The region's growing pharmaceutical sector, particularly in China, stands to benefit from AI-driven efficiency, but must also address the bottlenecks of human biology and clinical execution. Our view is that the US$26 billion in potential cost savings from AI-powered drug discovery is a strong incentive, but companies should focus on integrating AI as an accelerant rather than a standalone solution. The challenge for Asian companies will be to translate AI's speed into tangible profits and patient benefits, as noted by John Yung of Citi. This requires a nuanced approach that combines AI's data processing power with deep human biological and clinical understanding.

    Original reporting by SCMPWe don't republish, read the full story â†’

    Related reading

    6 stories