The $2 Billion Bet: Why Isomorphic Labs (Google) Is Raising Massive Capital for AI Drug Discovery

On May 8, 2026, Bloomberg reported that Isomorphic Labs — the AI drug discovery company spun out of Alphabet's Google DeepMind — is in advanced discussions to raise more than $2 billion in a new funding round. The round is set to be led by Thrive Capital, the venture firm that also led Isomorphic's first external funding round just fourteen months ago. Alphabet itself is participating as a follow-on investor, and the round has not yet closed.

The number is striking on its own terms. But its significance runs deeper than the headline figure. This raise — if completed — would bring Isomorphic Labs' total external capital to over $2.6 billion in under two years, making it one of the most heavily capitalised AI life sciences ventures in history. It comes at a pivotal moment: the company's first AI-designed drugs are preparing to enter human clinical trials, its internal drug design engine has reportedly surpassed AlphaFold 3 on key benchmarks, and it carries active research partnerships with pharmaceutical giants Eli Lilly, Novartis, and Johnson & Johnson carrying combined potential milestone payments approaching $3 billion.

This article explains what Isomorphic Labs is, why it is raising at this scale and at this moment, what the money is expected to fund, and why the entire question of whether AI can actually redesign the economics of pharmaceutical drug development is now being tested in real patients rather than research papers.

What Is Isomorphic Labs?

Isomorphic Labs was founded in 2021 as a commercial spinout from Google DeepMind — the London-based AI research laboratory that has produced some of the most consequential scientific AI systems in history, including AlphaFold. The company is headquartered in London with a second office in Lausanne, Switzerland.

Its founding mission is stated plainly and ambitiously: to transform drug discovery with the power of artificial intelligence, with a long-term goal of one day solving all disease with the help of AI.

The founding intellectual asset is AlphaFold — DeepMind's AI system that solved the fifty-year-old protein folding problem: the challenge of predicting the precise three-dimensional structure of a protein from its amino acid sequence. Knowing a protein's structure is foundational to drug design, because most drugs work by binding to specific protein targets in the body. If you can accurately predict how a protein is shaped, you can begin designing molecules that fit into it like a key in a lock.

AlphaFold 2 predicted the structures of virtually all known proteins and published them in an open database — an act of scientific generosity that the global research community immediately used to accelerate work across biology and medicine. AlphaFold 3, co-developed by Isomorphic Labs and Google DeepMind and released in May 2024, extended the capability further: it can now predict the structure and interactions of all of life's molecules, including proteins, small molecule drugs, DNA, and RNA simultaneously. In January 2024, Sir Demis Hassabis — who leads both Google DeepMind and Isomorphic Labs — and colleague John Jumper were awarded the 2024 Nobel Prize in Chemistry for AlphaFold.

Isomorphic Labs was established to take that foundation and build the commercial infrastructure needed to translate it into actual medicines.

The Technology Stack: From AlphaFold to IsoDDE

AlphaFold provided the map of biological structure. Isomorphic Labs is building the vehicles to navigate it.

The company's core proprietary technology is the Isomorphic Labs Drug Design Engine (IsoDDE) — an internal AI system that extends beyond structure prediction into de novo drug design: not merely predicting how existing molecules interact with protein targets, but generating entirely new molecular structures with desired binding properties.

IsoDDE provides predictive accuracy for molecular interactions, enabling rational drug design across multiple therapeutic areas and drug modalities. The practical implication is that IsoDDE can screen millions of candidate molecules in a digital environment before a single compound is synthesised in a laboratory. Traditional wet lab drug discovery works in roughly the reverse order: synthesise compounds, test them experimentally, observe failures, and iterate. The cost and time of that iteration cycle — multiplied across thousands of candidate compounds over years — is a primary driver of the industry's famously high development costs and low success rates.

In February 2026, Isomorphic Labs released a technical report showing that IsoDDE had surpassed AlphaFold 3 on key benchmarks for drug design accuracy. Max Jaderberg, the Chief AI Officer, described the results as showing "huge progress in accuracy and capabilities critical for in-silico drug discovery." Hassabis noted the results publicly: the drug design engine is extending the state of the art further across key benchmarks. The shelf life of a Nobel Prize, in this case, appears to be measured in months — AlphaFold's record was broken by the company built on it.

The therapeutic strategy concentrates initial programmes in oncology and immunology — disease areas where protein structure understanding is most directly actionable and where unmet clinical need is highest. The company has also targeted "un-druggable" proteins: targets whose shapes were previously too complex or elusive to be addressed by conventional pharmaceutical methods. AlphaFold's ability to model these structures opens drug design possibilities that simply did not exist before structural biology at this accuracy was possible.

The Funding History: From DeepMind Spinout to $2 Billion+

Understanding the scale of the current raise requires context on how quickly Isomorphic's financing has escalated.

2021–2024: Alphabet-funded research phase. From its founding through early 2025, Isomorphic Labs operated entirely on Alphabet's balance sheet. No external investors. No disclosed funding rounds. The company was effectively an internal R&D bet within one of the world's most cash-rich technology companies.

March 2025: $600 million Series A. Isomorphic raised its first external funding — $600 million led by Thrive Capital, with participation from GV (Google Ventures) and Alphabet. Thrive Capital founder and CEO Joshua Kushner called Isomorphic a "category-defining company." The $600 million was designated to advance the AI drug design engine, progress internal drug candidates toward clinical development, and scale the team. At the time, Hassabis stated that the funds would "turbocharge" AI development toward solving diseases with AI.

May 2026: $2 billion+ Series B (in advanced discussions). Fourteen months after the Series A, Isomorphic is in advanced discussions for a round more than three times larger. Thrive Capital is again leading — a strong signal of continued conviction from its primary backer. Alphabet is participating again. The round has not yet closed, and final terms have not been disclosed.

The 3.3x increase in round size in fourteen months implies a significant multiple expansion in Isomorphic's valuation and reflects both the general acceleration of capital flows into AI life sciences and Isomorphic's specific progress in advancing its programmes toward clinical readiness.

The sheer size of the new funding round implies a major multiple expansion from Isomorphic's initial $600 million debut.

Why $2 Billion and Why Now? The Three Drivers

The timing and scale of this raise are not arbitrary. Three converging factors explain both.

Driver 1: Clinical Trials Require a Different Capital Profile

The transition from computational drug design to human clinical trials is the most capital-intensive inflection point in pharmaceutical development. It is also called the "valley of death" in biotech — the stage where computational predictions must survive contact with the unpredictable complexity of human biology, and where most drug candidates historically fail.

Phase 1 clinical trials — the first-in-human studies that test safety and dosing — cost tens of millions of dollars per candidate. Phase 2 trials, which test efficacy in a targeted patient population, cost hundreds of millions. Running multiple candidates through parallel clinical programmes simultaneously — the strategy that maximises the probability of producing at least one approved drug — requires the kind of capital that no Series A round, however large, fully provides.

Isomorphic Labs was looking to advance AI-designed drug candidates to clinical stages, with plans to develop candidates through early-stage trials before licensing them. That licensing model — advancing drugs to early clinical proof-of-concept and then partnering with large pharma for later-stage development — is capital-efficient compared to running full development programmes independently. But it still requires sustained investment through Phase 1 and into Phase 2 before a licensing deal can be structured at commercially meaningful terms.

The $2 billion raise provides the runway to run multiple clinical programmes simultaneously without depending on partnership milestones to fund operations — a position of significantly greater strategic leverage than operating on partnership income alone.

Driver 2: The Clinical Trial Timeline Reset

The urgency of the capital raise is partly a response to a timeline that slipped.

CEO Demis Hassabis delayed the company's first clinical trials from the end of 2025 to the end of 2026. At the World Economic Forum in Davos in January 2025, Hassabis had stated publicly that Isomorphic hoped to have its first AI-designed drug in clinical trials by the end of 2025. By September 2025, he described showing "the first few proof points" for moving drugs to trial but declined to give a specific timeline. The company subsequently reset the target to end of 2026.

This delay is not unusual in biotech — clinical programme timelines routinely shift by months or years as the complexity of regulatory preparation, candidate selection, and trial design is worked through. But it does mean the company is entering this fundraise in a market that has moved from excitement about AlphaFold's theoretical promise to demanding clinical validation. The $2 billion provides not just the capital to run the trials but the financial stability to absorb the timeline uncertainty that clinical development inherently involves.

The massive new funding round is a bet that the company can now meet this reset timeline. But the expectation gap itself — between the sector's early promise and the slow grind of clinical validation — remains a significant overhang.

Driver 3: Alphabet's Commercialisation Strategy

The third driver is strategic rather than operational. A larger raise would suggest Alphabet is preparing Isomorphic for a much broader role than partnership work alone.

Alphabet has invested in AI capabilities across multiple domains — autonomous vehicles through Waymo, life sciences through Verily, quantum computing through Google Quantum AI, and foundational AI through DeepMind. Most of these ventures have been long-incubated on Alphabet's balance sheet before raising external capital. Isomorphic's shift from entirely internal funding to large external rounds follows a pattern: establishing proof of concept internally, then raising external capital to scale the commercial operation independently.

The $2 billion raise accelerates Isomorphic's separation from pure internal research status into an independently-capitalised commercial biopharmaceutical company. The participation of Thrive Capital — a top-tier venture firm with extensive biopharmaceutical investment experience — and continued Alphabet backing signals a dual mandate: independent commercial validation and parent-company strategic alignment.

Alphabet is not simply funding a research unit. It is building a new category of pharmaceutical company — one where the primary competitive advantage is AI-generated intellectual property about molecular structure and drug-target interactions, rather than synthetic chemistry or clinical expertise inherited from legacy pharmaceutical operations.

The Partnership Portfolio: $3 Billion in Potential Milestones

The $2 billion raise lands on top of a partnership portfolio that already represents significant validation from the conventional pharmaceutical industry.

Eli Lilly: One of the world's largest pharmaceutical companies, with a market capitalisation exceeding $700 billion and a dominant position in diabetes and obesity treatments. The Isomorphic-Lilly collaboration is focused on advancing drug programmes using Isomorphic's AI platform, with potential milestone payments forming a significant portion of the combined approximately $3 billion in potential milestone payments across Isomorphic's pharmaceutical partnerships.

Novartis: The Swiss pharmaceutical giant is Isomorphic's second major pharma partner, with a collaboration focused on applying IsoDDE to Novartis's drug discovery pipeline across multiple therapeutic areas. Novartis has been one of the pharmaceutical industry's most active investors in AI-assisted drug discovery, making it a natural fit for a platform like IsoDDE.

Johnson & Johnson: The third major pharmaceutical partner, further broadening the therapeutic area coverage of Isomorphic's collaborative programmes.

These partnerships serve two distinct functions. Commercially, they provide near-term revenue and milestone payments that partially fund operations. Strategically, they provide external validation that the world's largest pharmaceutical companies — with their own internal AI capabilities, their own deep structural biology expertise, and their own rigorous scientific review processes — believe Isomorphic's platform is superior enough to their own approaches to warrant paying for access.

Partnership deals with Lilly, Novartis, and J&J are not marketing endorsements. They are procurement decisions made by organisations that have spent billions of dollars building their own drug discovery capabilities. When they pay a premium to access Isomorphic's platform, they are communicating a specific assessment: that IsoDDE produces outputs their own systems cannot reliably match.

The Traditional Drug Discovery Problem IsoDDE Is Designed to Solve

To understand why $2 billion is a rational investment in this platform, you need to understand the economics of the problem it is attacking.

Traditional drug discovery faces three fundamental challenges that compound into an industry-wide efficiency crisis:

The time problem. Developing a new drug from initial target identification to regulatory approval takes an average of 10 to 15 years. The majority of that time is spent in iterative experimental cycles — synthesise a compound, test it, observe the failure mode, redesign, repeat. Hassabis has articulated a target of compressing that timeline to months rather than years using computational prediction to front-load the decision-making.

The cost problem. The fully-loaded cost of bringing a single new drug to market — accounting for the failures along the way — has been estimated at over $2 billion per approved drug by several independent studies. The primary driver is the high attrition rate: most drug candidates fail in clinical trials, and the cost of running those failed trials must be amortised across the few that succeed. If AI can improve the pre-clinical prediction accuracy — selecting better candidates before the expensive clinical phase — the attrition rate drops and the effective cost per successful drug drops with it.

The target problem. Approximately 85% of disease-relevant proteins have been considered "un-druggable" by conventional methods — either because their structures were unknown, too flexible to target with small molecules, or too similar to other proteins to target selectively. AlphaFold and AlphaFold 3 have fundamentally changed the structural biology of many of these targets, opening design opportunities that did not previously exist. IsoDDE then provides the computational tool to design molecules that exploit those newly revealed structures.

As Hassabis noted at Davos in 2025, Isomorphic is working on shortening the drug discovery process from a decade or more to weeks or months. Whether that compression survives contact with clinical reality — with the regulatory process, with human biology's complexity, with the unpredictability of adverse effects that no computational model fully anticipates — is the question the current clinical programme is designed to answer.

What the $2 Billion Will Fund

The new funding round will be used to strengthen Isomorphic Labs's drug design engine and support its global expansion. That description, from NewsBytesApp citing people familiar with the plans, covers three operational categories.

Clinical programme advancement. Running multiple Phase 1 trials simultaneously in oncology and immunology — the therapeutic areas where Isomorphic's internal pipeline is concentrated — requires sustained capital at a level that the $600 million Series A was not sized to provide indefinitely. The $2 billion provides a clinical runway of several years at the programme intensity required to generate meaningful proof-of-concept data.

Platform development. IsoDDE is not a finished product. It is a continuously evolving AI system that improves as it is exposed to more molecular data, more protein structures, and more experimental validation results from its own programmes. The capital funds the engineering team, computational infrastructure, and data acquisition required to keep IsoDDE ahead of both internal expectations and external competition.

Global expansion. Isomorphic operates from London and Lausanne. Expanding into additional geographies — particularly the United States, where the largest pharmaceutical market and the greatest concentration of oncology clinical infrastructure are located — requires capital for facilities, team, and regulatory infrastructure.

Team scaling. Isomorphic has recruited heavily from Google DeepMind, building a team that combines deep machine learning expertise with structural biology knowledge. Scaling that combination — which is genuinely rare in the broader talent market — requires competitive compensation that a pre-revenue biotech cannot sustain without substantial capital reserves.

The Honest Risk Assessment

Intellectual honesty about Isomorphic Labs requires acknowledging the genuine risk factors alongside the genuine promise.

No AI-designed drug has yet completed a clinical trial. This is the most important fact in the entire AI drug discovery sector. The fundamental premise of the entire industry — that AI-generated drug candidates will survive clinical validation at higher rates than conventionally designed drugs — has not yet been proven. Clinical trials are the test, and the test has not been scored yet.

Traditional drug development faces around a 10% success rate once trials begin — the 90% failure rate is the primary driver of the industry's high costs. If AI-designed drugs fail at the same rate, the economic case collapses. If they succeed at meaningfully higher rates, the economic case for massive AI drug discovery investment is validated. Isomorphic's trial results — when they emerge — will be among the most consequential data points in the history of pharmaceutical investment.

The transition from computational models to physiological reality is not guaranteed. Transitioning from a digital environment to the clinic is the "valley of death" for many biotech startups. Computational prediction of protein-ligand binding affinity is a different problem than predicting clinical efficacy and safety in a human being — a system of extraordinary complexity operating under conditions that no model fully captures. IsoDDE may produce excellent binders that prove toxic, poorly bioavailable, immunogenic, or simply ineffective in the complexity of a living patient.

Timeline reliability is an open question. The delay from the original end-of-2025 clinical trial target to end-of-2026 is a single data point, but it reflects a broader pattern in AI biotech: the gap between computational achievement and clinical readiness is longer than optimistic projections suggest. Managing expectations about how quickly AI-designed drugs will reach patients is an ongoing challenge for the entire sector.

Competition is intensifying. Isomorphic is not operating in an uncontested space. Insilico Medicine, Recursion Pharmaceuticals, Exscientia, AbSci, BioNTech's AI division, and several stealth-mode programmes at large pharmaceutical companies are all pursuing variations of the same opportunity. The specific combination of AlphaFold's structural biology foundation, IsoDDE's design capability, and Isomorphic's pharmaceutical partnerships represents a real competitive advantage — but it is not an insuperable moat in a domain where computational biology is advancing rapidly across many organisations.

Why This Matters Beyond the Investment

The Isomorphic Labs $2 billion raise is not primarily a financial story. It is a signal about where the pharmaceutical industry believes the next decade's drug pipeline will come from.

When Thrive Capital leads a round of this size — at a company that has yet to produce a single approved drug — it is making a statement about what drug discovery will look like at scale by 2030. When Alphabet participates despite already funding the company internally for five years, it is doubling down on a long-term thesis about the commercial monetisation of DeepMind's scientific work. When Eli Lilly, Novartis, and Johnson & Johnson pay for access to IsoDDE despite having their own internal AI capabilities, they are hedging against the possibility that external AI platforms will outpace internal development.

The bet is simultaneously that AI will make drug discovery faster and cheaper, that Isomorphic Labs specifically has built the best AI platform for doing so, and that the clinical results now being generated will validate enough of the computational predictions to justify the investment.

If Isomorphic succeeds — if AI-designed drugs move through Phase 1 with better safety profiles than expected, if Phase 2 results demonstrate efficacy that computational models predicted, if the timeline compression from target identification to clinical candidate happens in months rather than years — the implications extend far beyond a single company's returns. It would validate computational pharmacology as a primary mode of drug development, accelerate the deployment of AI across the entire pharmaceutical industry, and potentially begin to change the economics of bringing medicines to patients in ways that affect healthcare costs at a societal scale.

If the clinical trials produce failures — if the computational predictions do not survive contact with human biology at the rates the investment implies — the market for AI drug discovery will face a significant confidence reset, and the capital currently flowing into the sector will reprice accordingly.

The trials are running. The results will come. And the $2 billion is, ultimately, a bet on what those results will say.

Quick Reference: Isomorphic Labs at a Glance