(Bloomberg) — Discovering new medicines takes years and is marred by an astronomical failure rate. The Nobel Prize-winning head of Alphabet Inc.’s artificial intelligence lab says the technology will soon trim that time to under a year.
“In the next couple of years, I’d like to see that cut down in a matter of months, instead of years,” Demis Hassabis said in an interview with Bloomberg Television. “That’s what I think is possible. Perhaps even faster.”
Hassabis runs Google DeepMind, the company’s AI unit, as well as Isomorphic Labs, an Alphabet division focused on drug discovery. Since forming in 2021, Isomorphic Labs has cut deals with pharmaceutical companies Eli Lilly & Co. and Novartis AG.
Pharmaceutical and biotech companies have touted AI-powered drug discovery as a way to speed up patient access to new treatments, cut development costs and respond to health-care crises more quickly. The field has shown major advances in computing, using algorithms to grasp large swaths of molecular data. But no AI-designed drugs have completed a successful clinical trial, meaning none have made their way to patients receiving care.
In January, Hassabis said his company would begin clinical trials of AI-designed drugs by the end of the year.
It hasn’t yet. Speaking from his office in London in early September, Hassabis said the company has shown the “first few proof points” for delivering drugs to trial but didn’t provide an update on timing. “It’s a bit early to say,” he said.
Isomorphic Labs was created to commercialize AlphaFold, DeepMind’s AI system that predicts protein behavior. Hassabis and another DeepMind scientist, John Jumper, shared the 2024 Nobel Prize in chemistry with a US professor for their research.
Hassabis said his researchers are working on a “much more advanced” version of the latest AlphaFold model, capable of understanding more than just protein interactions.
He has previously said Isomorphic Labs could build a business worth more than $100 billion. Earlier this year, it raised a $600 million financing round led by Thrive Capital.
The Alphabet unit is working to find treatments for cancer and immune disorders. Rebecca Paul, its director for medicinal drug design, said the diseases offer a relatively simpler path for translating work on algorithmic models into clinical results.
AI-developed drugs will turn many cancers into a treatable chronic disease, Paul said in a separate interview with Bloomberg Television. “It’s really difficult to put a time stamp on it,” she said. “But we can start to think about how we can approach this problem now.”
Last year, Isomorphic Labs said it was working with Novartis to discover therapeutics based on three targets, which are the proteins or molecules that a drug is designed to interact with. Hassabis said the companies are now focused on six targets.
Paul said there was “really good progress,” in the partnership between the companies without providing details.
More stories like this are available on bloomberg.com
About the Author
