跳到主要内容
Learn Motion
探索课程产品能力价格

下载 Learn Motion

用手机扫描二维码,下载 App。

App Store

扫码从 App Store 下载 Learn Motion打开商店页面

Google Play

扫码从 Google Play 下载 Learn Motion打开商店页面
登录
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
Where AI Actually Helps in the Pipeline

Why the help thins out as the pipeline moves forward

3 / 3
Watch the two quantities move in opposite directions. At the start, data is plentiful and each additional data point costs almost nothing, because the structures and measurements already exist in databases. As the flow moves forward, the data bar shrinks and the cost bar grows: experiments in cells and animals have to be run and paid for, and human trials take years and depend on patients. The animation is showing you the cause of the shading you saw earlier. AI is not stronger at the front because the problems there are easier; it is stronger there because that is where the data is.
0:00 / 0:00

The front-loaded pattern has a single underlying cause: the kind of data each stage produces.

At the discovery and design stage, the working material is already digital. Protein structures, chemical structures, and measured properties live in databases, some of them public and containing millions of entries. A model can be trained on that material and can be retrained as more is added. The cost of one more data point is close to zero, because the data was collected for other reasons and is simply being reused.

Preclinical testing changes the picture. Here the data comes from experiments in cells and animals, and each experiment has to be designed, run, and paid for. The volume is smaller and the cost per data point is much higher. AI can still help — for example, by predicting which compounds are worth testing first — but it is working with less material and cannot generate more of it on demand.

Clinical trials change it again. The data comes from human volunteers and patients, and it arrives over years, in limited numbers, under strict rules. You cannot run a trial faster by writing better software, because the constraint is the availability of people and the time it takes for a disease to progress. Regulatory approval and post-approval monitoring sit at the far end of the same gradient: the evidence is produced by clinicians and patients, and it accumulates slowly.

So the gradient in the diagram is really a data gradient. Where data is abundant and cheap, AI has room to work. Where data is scarce and expensive, its influence shrinks — not because the later stages matter less, but because there is less for a model to learn from.

上一页3 / 3下一页

Learn Motion

生成课程,认真学会。

运营方:Wuhan Daoyin Technology Co., Ltd.

联系我们: [email protected]
隐私政策用户服务协议

© 2026 Learn Motion