AI glossary · Basic · Science & bio · Healthcare & bio
AI drug discovery
AI drug discovery uses machine learning and generative AI to find targets that cause disease, design and screen substances that could become drugs, and predict their effects and side effects in advance, making drug development faster and more efficient.
About this term
- Level
- Basic · Technical terms you often meet in AI news
- Category
- Industry AI
- Fields
- Science & bio
- Industries
- Healthcare & bio
- Added
- 2026-10-10
How it works
Drug development is a long process: choosing a "target" such as a disease-related protein, finding a substance that binds to and acts on that target, then verifying safety and efficacy through animal studies and clinical trials. AI analyzes genomic and protein data to find promising targets and predicts protein structures to estimate where a drug will bind. Generative models are also used to design new molecular structures or to virtually screen huge compound libraries for likely effective candidates. AI is also used to predict toxicity and absorption in the body in advance, reducing the number of candidates to test.
How candidates narrow down
Why it matters
Traditional drug development takes a long time and costs a lot, and most candidate substances fail along the way. There are high hopes that if AI filters out unlikely candidates early, the number of experiments and the time needed can be reduced. Another strength is that it can propose molecules with novel structures that people would find hard to think of. That is why large pharmaceutical companies, big tech firms and startups are all entering this field.
Things to know
Candidates proposed by AI must still go through the lab and clinical trials, and the risk of failure at the clinical stage remains high. Good predictions require large amounts of high-quality biological data, but such data is expensive and scattered across companies. That is why joint projects that collect cell and protein data at scale are becoming important. The phrase "AI-made drug" can mean very different things depending on which stage AI contributed to and how much.
Example
News on this site included a report that Google DeepMind, Meta and Isomorphic Labs are jointly investing in building "virtual cell" data led by Biohub, the organization backed by Mark Zuckerberg and Priscilla Chan. Building models that simulate on a computer how cells respond to drugs or genetic changes requires large-scale data, and such data forms the foundation of AI drug discovery. Companies that specialize in finding drug candidates with AI, such as Isomorphic Labs, are leading this trend.