Live·Open questions in longevity research
All news
Biotech fundingAI in medicineTherapeutics

Demis Hassabis’s Isomorphic Labs raises $5 billion for AI drug design and preparations for its first clinical trials

18 November 2026· 261118001

Demis Hassabis’s Isomorphic Labs raises $5 billion for AI drug design and preparations for its first clinical trials

On November 17, Isomorphic Labs announced its Series C round, its third external funding round. Greenoaks and Sequoia Capital led the round, with Andreessen Horowitz among the new investors. Alphabet, Isomorphic’s parent company, and all previous investors also participated. Isomorphic did not disclose its valuation; Bloomberg reported on November 4, citing sources, that talks were taking place at a valuation of about $50 billion.

The company raised $600 million in March 2025 and $2.1 billion on May 12, 2026. The latest $5 billion brings the total to about $7.7 billion. According to BioSpace, the Series B was the second-largest funding round in biotech history, and the Series C set a new record. The previous record was the $3 billion raised at the 2022 launch of Altos Labs, a company studying cellular rejuvenation.

Isomorphic was spun out of Google DeepMind in 2021 to build on AlphaFold, a model that predicts protein structures. Drug design also requires predictions of which molecules will bind to a protein and how tightly they will bind. IsoDDE, a proprietary drug design engine, combines these predictions. In February, scientists quoted in Nature compared the scale of the advance to “AlphaFold 4”, while developers of open models speculated about how to reproduce the result.

Company president Max Jaderberg described how the engine works in September. Drug designers specify a target (usually a protein the drug is intended to act on), a molecule type, and the desired properties. An AI agent then autonomously generates successive batches of molecules and evaluates them using model predictions alone. In the example he gave, the search took two to four days. Traditionally, Jaderberg writes, each cycle of this search takes one to three months: chemists design and synthesize molecules, biologists test them, and the resulting data are analyzed. Once the search is complete, a handful of molecules are synthesized and tested in the laboratory instead of thousands or millions. In his example, the molecule identified by the agent acted on the target.

Sequoia partner Konstantine Buhler explained to Bloomberg why the firm is confident that Isomorphic will reach clinical trials. The company has identified scaling laws for its models, meaning that their performance improves predictably with computing power and data volume:

That matters a great deal because you at least know where you are going. You have a well-informed basis for committing capital and resources.

Company founder Hassabis describes the goal this way:

We are trying to build a platform that can create hundreds of drugs a year. The idea is to focus on the technology and platform needed to do this at scale, rather than on any particular program or drug.

The funding will support further development of IsoDDE and AI models of biology, a network of partners studying disease mechanisms, and the company’s own clinical infrastructure. Jaderberg describes the company as vertically integrated: it develops models, generates data, and will take drug candidates into the clinic for testing in patients. Hassabis also wants to cover “virtually all modalities,” meaning types of drugs, including large-molecule biologics.

The company says its internal oncology and immunology programs, along with its partner programs, have produced positive preclinical results. Hassabis added that more than 20 programs are underway. In January, he expected the first clinical trials to begin by the end of 2026. In the November interview, he gave no updated timeline, but said regulatory submissions were on schedule and that he would share more in 2027.

Originally published on Telegram by Ukhvat NewsView on Telegram
Sources
#isomorphic-labs#ai-drug-design#series-c#isodde#alphafold#clinical-pipeline