From AlphaFold to AI-powered labs, artificial intelligence is accelerating drug discovery and reshaping the future of medicine and biotechnology.
When AlphaFold2 stunned the scientific community in 2020 by accurately predicting the shapes of proteins, it was hailed as a breakthrough that could transform biology forever. Developed by DeepMind, the artificial intelligence system solved a challenge that had taken scientists decades to tackle through painstaking laboratory work.
The impact was immediate. AlphaFold went on to model more than 200 million proteins, earning its creators, Demis Hassabis and John Jumper, the 2024 Nobel Prize. Many believed the technology would rapidly accelerate drug discovery and potentially unlock treatments for some of humanity’s most challenging diseases.
Yet five years later, the much-anticipated wave of AI-designed medicines has not fully arrived.
The reason is simple: biology is far more complex than predicting the shape of a single protein.
Proteins are constantly moving, interacting with DNA, RNA and countless other biological molecules in a highly dynamic environment. Understanding these interactions requires much more than a static snapshot. Scientists must still determine how molecules behave over time and how they influence one another inside living systems.
As a result, the field has evolved significantly beyond AlphaFold.
New generations of AI tools such as RoseTTAFold and advanced biological language models are now capable of modelling proteins, DNA, RNA and small molecules simultaneously. Researchers are even developing systems that can design entirely new proteins, generate functional genomes and create novel antibodies that are already entering clinical trials.
Perhaps most significantly, AI is beginning to move beyond modelling and into scientific research itself.
Platforms such as FutureHouse’s Robin and DeepMind’s Co-Scientist are designed to analyse scientific literature, generate hypotheses, recommend experiments and help interpret results. Combined with advances in robotics, AI systems are increasingly capable of automating parts of the research process that once required teams of scientists.
Earlier this year, OpenAI and Ginkgo Bioworks revealed that GPT-5 had been used to independently design and execute thousands of biological experiments in a cloud-based robotic laboratory. The project reportedly reduced the cost of producing a target protein by 40 percent, highlighting how AI could make scientific discovery faster and more efficient.
The potential benefits are enormous.
Drug development currently takes an average of 10 years and billions of dollars to bring a single treatment to market. AI could dramatically shorten the early stages of discovery by identifying promising compounds, predicting outcomes and reducing the number of failed experiments.
Some experts believe these advances could eventually lead to a future where new medicines are developed in months rather than years.
However, significant challenges remain.
Clinical trials still require years of testing to ensure safety and effectiveness. No matter how sophisticated AI becomes, there are no shortcuts around the rigorous regulatory processes needed to protect patients. Furthermore, many promising drugs fail during testing, regardless of how they were designed.
There are also growing concerns about misuse. As AI lowers the barriers to complex scientific tasks, experts warn that powerful biotechnology tools could become accessible to individuals without specialised training. While safeguards exist, researchers agree that stronger oversight will be needed as AI capabilities continue to advance.
Despite these concerns, the overall trajectory remains encouraging.
Rather than replacing scientists, AI is becoming a powerful partner in discovery, helping researchers analyse data faster, generate new ideas and explore scientific questions at unprecedented scale.
The dream of curing every disease may still be far away, but AI is already transforming biology from a slow, labour-intensive process into a faster, more computationally driven discipline. While the technology may not deliver miracles overnight, it is steadily changing how science is done and bringing humanity closer to breakthroughs that once seemed impossible.
