DeepMind’s isomorphic lab has the epic ambition to “cure all diseases” with AI
Alphabet’s secret drug discovery arm, Isomorphic Lab, is preparing to begin testing AI-designed drugs with Humans, Colin Murdoch and President of Isomorphic Labs. Google Deepmind’s chief business officer said luck.
“There’s someone sitting in our office in King’s Cross, London. Interview in Paris. “It’s happening now.”
A few years in development, Murdoch says human clinical trials of the isotype AI-assisted drug are finally visible.
“The next big milestone was actually going out to clinical trials and starting to put these things in humans,” he said. “We’re on staff now. We’re very close.”
The company that was spun deepmind It was born in 2021 from one of Deepmind’s most famous breakthroughs, Alphafold. This is an AI system that can predict protein structure with high levels of accuracy.
The interaction of Alphafold progressed to the point that it was able to accurately predict individual protein structures to model how proteins interact with other molecules such as DNA and drugs.
These leaps have made them much more useful for drug discovery, helping researchers design drugs faster and more accurately, and have transformed the tool into launchpad for much greater ambitions.
“This was the inspiration for the isomorphic lab,” Murdoch said of Alphafold. “It really shows that you can do something very basic with AI that helps you unlock drug discovery.”
Major research collaborations with the equally signed Pharma Companies, which released Alphafold 3 in the same year in 2024. Novartis and Eli Lily.
A year later, in April 2025, Isomorphic Labs raised $600 million in the first-ever external funding round led by Thrive Capital.
The deal is part of Isomorphic’s plan to build a “world-class drug design engine.” This is a system that combines machine learning researchers with Pharma veterans to design new drugs faster, cheaper, and more likely to succeed.
As part of its deal with major pharmaceutical players, the isomorph supports existing drug programs, but aims to design unique internal drug candidates in fields such as oncology and immunology, and ultimately license them after early stage testing.
“We identify unmet needs, launch our own drug design program. We develop them and put them in human clinical trials… We don’t have that yet, but we’re making good progress,” he said.
Today, pharma companies often spend millions trying to bring a single drug into the market, but once trials begin, there was only a 10% chance of success. Murdoch believes that Isomorphic’s technology can fundamentally improve these odds.
“We’re trying to do all of this, speeding up them, reducing costs, but it really improves the chances of success,” he says. He wants to use Alphafold’s technology to reach a point where he has 100% certainty that the drugs he is developing will work in human testing.
“One day, we hope we can say. Well, here’s the illness. Then click the button and pop the drug design to deal with that illness,” Murdoch said. “All of these come with these incredible AI tools.”