Skip to content
Learn Motion
ExploreHow it worksMembership
Log in
Learn Motion

Why AI-Designed Drugs Haven't Changed Medicine Yet

1The Promise and the Puzzle2From Molecule to Medicine: The Journey a Drug Must Survive3Where AI Actually Helps in the Pipeline4The Prediction Gap: When a Good Molecule Meets a Real Body5The Long, Expensive Road of Clinical Trials6Money, Incentives, and the Business of Drug Development7Regulation, Evidence, and Trust8What Would Have to Change
The Long, Expensive Road of Clinical Trials

Why a Faster Start Does Not Mean a Faster Finish

4 / 4
The comparison on this page is the heart of the argument. On the left are the things that can be accelerated: molecule design and data analysis, because they run on computation and can be parallelized. On the right are the things that cannot: preclinical safety testing, recruitment, trial follow-up, event accumulation, and regulatory review. Each of those is gated by a different clock — animal biology, human willingness, how long a body takes to respond, and how long it takes to gather enough evidence. Now look at the example. If design time falls from three years to three months, but the rest of the path still takes about ten years, the total goes from roughly thirteen years to about ten and a quarter. That is a real saving, but it is modest, and it does not change the fact that most of the wait happens after the molecule already exists. Speeding up one gate does not open the others.
0:00 / 0:00

Faster molecule design shortens only the discovery stage, which is a small fraction of the total timeline. The remaining stages are gated by biological and human constraints that computation cannot compress.

What gates each stage

Can be accelerated

  • Molecule design: computation and chemistry, parallelizable
  • Data analysis: software scales with hardware

Cannot be accelerated by AI

  • Preclinical safety testing: animal biology and assessment time
  • Recruitment: eligibility, geography, willingness
  • Trial follow-up: how long a body takes to respond
  • Event accumulation: waiting for enough clinical outcomes
  • Regulatory review: careful examination of evidence

Suppose design time falls from three years to three months. If the rest of the path — preclinical work, trial setup, recruitment, the trials, and review — still takes about ten years, the total goes from roughly thirteen years to about ten and a quarter. The saving is real but modest, and it does not change the fact that most of the wait happens after the molecule exists.

Previous4 / 4Next

Learn Motion

Generate a course. Learn it properly.

Operated by Wuhan Daoyin Technology Co., Ltd.

Contact: [email protected]
Privacy PolicyTerms of Service

© 2026 Learn Motion