AI-Designed Drugs by DeepMind Spinoff Near Human Trials
Isomorphic Labs Moves AI-Designed Drugs Closer to Human Trials
Isomorphic Labs, the London-based AI drug discovery company founded as a commercial subsidiary of Alphabet and spun out of Google DeepMind, is preparing to enter its first Phase I clinical trials in 2026. The company, which has built a pipeline of drug candidates targeting oncology and immunological diseases, has signaled it is actively staffing up for the transition from computational design to human testing — a milestone that would mark a significant moment for the broader field of AI-driven drug discovery.
The move comes after a period of rapid technological development, major pharmaceutical partnerships, and a landmark $600 million fundraise. It also follows an acknowledged delay: founder and CEO Sir Demis Hassabis had previously stated at the World Economic Forum in Davos in January 2025 that AI-designed drugs could enter clinical trials by the end of that year. By January 2026, he acknowledged the timeline had slipped to the end of 2026. As of early 2026, no Isomorphic Labs drug had yet entered human clinical trials.
A New Drug Design Engine and a Pipeline in Progress
Central to Isomorphic Labs' clinical ambitions is its next-generation computational platform, the Isomorphic Labs Drug Design Engine, known as IsoDDE, which the company released on February 10, 2026. According to the company's own technical report, IsoDDE more than doubles the accuracy of AlphaFold 3 on a challenging protein-ligand structure prediction benchmark — a key measure of how precisely an AI model can predict how a drug molecule will bind to its biological target.
IsoDDE is described by the company as a unified computational drug-design system that goes beyond structure prediction to include capabilities such as binding affinity estimation and novel pocket discovery — features that could allow researchers to identify previously unknown sites on proteins where drugs might take effect. Max Jaderberg, who became President of Isomorphic Labs in January 2026 after previously serving as Chief AI Officer and a founding team member, has described the platform's underlying models as fundamentally distinct from other approaches in the field.
"The models behind IsoDDE are 'profoundly different' from other efforts," Jaderberg said, as quoted by Scientific American. He has also pointed to a familiar recipe driving the advances: "Like with most big machine-learning and AI advancements, it's a combination of compute, data [and] algorithms."
In his previous role as Chief AI Officer, Jaderberg led the research, development, and application of the AI models underpinning Isomorphic Labs' drug design work, including AlphaFold 3, which was co-developed with Google DeepMind and released in May 2024. The development of AlphaFold earned Sir Demis Hassabis and John Jumper the Nobel Prize in Chemistry in 2024.
Pharmaceutical Partnerships and a $600 Million War Chest
Isomorphic Labs has assembled a substantial commercial foundation to support its push toward clinical trials. In January 2024, the company signed research partnerships with both Eli Lilly and Novartis — deals that, according to the company, have the potential to be worth nearly $3 billion in total. The Eli Lilly agreement paid Isomorphic $45 million upfront, with a further $1.7 billion contingent on hitting certain development milestones. The Novartis contract carried $37.5 million upfront and a potential $1.2 billion in downstream milestone payments. The company also has a research collaboration with Johnson & Johnson.
In March 2025, Isomorphic Labs closed its first-ever external funding round, raising $600 million. The round was led by Thrive Capital, with participation from GV and follow-on capital from Alphabet. Sir Demis Hassabis framed the raise as a strategic alignment of expertise: "We're excited to bring together a top-tier investor group with deep AI and life sciences expertise as we aim to transform this industry through an interdisciplinary approach."
Joshua Kushner, Founder and CEO of Thrive Capital, offered a bullish assessment of the company's positioning: "We believe Isomorphic has earned a rare position to define a new age of drug discovery and design, and we are deeply inspired by their mission and the extraordinary progress they have made to date."
Beyond its partnered programs, Isomorphic Labs is also developing an internal pipeline of drug candidates focused on oncology and immunology — the same therapeutic areas it is prioritizing for its first Phase I trials.
Why This Moment Matters for AI Drug Discovery
Isomorphic Labs' anticipated move into clinical trials arrives at a pivotal moment for the AI drug discovery sector. According to GlobalData's Drugs database, there are currently more than 3,000 drugs developed or repurposed using AI, with most still in early stages of development. The field has attracted significant investment and scientific attention, but translating computational breakthroughs into approved medicines remains an enormous challenge.
The financial backdrop has also grown more cautious. Venture capital funding for AI drug discovery peaked at $1.8 billion in 2021 and cooled to roughly half that amount by 2025, according to Winbuzzer. Meanwhile, the clinical attrition rate for drugs in general remains daunting: only approximately 10% of drug candidates that reach clinical trials successfully navigate to regulatory approval, according to multiple sources including Winbuzzer and Fortune.
Against that backdrop, getting an AI-designed molecule into a Phase I trial — which tests safety in humans for the first time — would not guarantee commercial success, but it would represent a meaningful proof point for the technology's ability to produce viable candidates. It would also set Isomorphic Labs apart in a field where the gap between computational promise and clinical reality has often proven wider than anticipated.
The company's platform has drawn attention from the scientific community. Its IsoDDE system, built on the AlphaFold lineage of models, represents an attempt to create a fully integrated computational workflow for drug design rather than a collection of separate tools. The company maintains its primary headquarters in London's King's Cross, with an additional office at EPFL's Innovation Park in Lausanne, Switzerland, opened in May 2023.
What Leaders at Isomorphic Labs Are Saying
The urgency around clinical readiness has been a recurring theme in public statements from the company's leadership. "The next big milestone is actually going out to clinical trials, starting to put these things into human beings. We're staffing up now. We're getting very close," said Colin Murdoch, Chief Business Officer of Google DeepMind, as quoted by Fortune.
Murdoch has also offered a ground-level description of the work already underway: "There are people sitting in our office in King's Cross, London, working, and collaborating with AI to design drugs for cancer. That's happening right now."
The company was founded in 2021 by Sir Demis Hassabis with an explicit mission to reimagine drug discovery from the ground up using AI — an ambition he has described in terms of ultimately working to solve all disease. Whether that ambition can be translated into clinical outcomes remains to be demonstrated, but the company's entry into human trials, when it occurs, will be a closely watched test of that vision.
What Comes Next
Isomorphic Labs has stated its intention to enter Phase I clinical trials within 2026, with oncology and immunology as the priority therapeutic areas. The company's pipeline includes both internally developed drug candidates and programs conducted in partnership with Eli Lilly, Novartis, and Johnson & Johnson.
The IsoDDE platform, released earlier this year, is positioned as the computational backbone for these efforts. Whether the accuracy gains the company has reported on benchmark tests translate into meaningful improvements in clinical outcomes is a question that only human trials can begin to answer. The 2026 timeline that Sir Demis Hassabis acknowledged at Davos in January 2026 represents a revised target after the original end-of-2025 goal was not met, adding a degree of caution to forward-looking assessments.
For the broader AI drug discovery field, Isomorphic Labs' progress — and its setbacks — are likely to serve as a significant reference point. With more than 3,000 AI-touched drug candidates currently tracked across the industry, the race to demonstrate that AI can meaningfully accelerate the path from molecule to medicine is intensifying. Isomorphic Labs, with its Nobel Prize-winning scientific heritage, deep-pocketed backers, and integrated computational platform, is among the most closely watched participants in that race.
For more tech news, visit our news section.