Alphabet's Isomorphic Labs, the pharmaceutical AI spinoff from Google DeepMind, is raising more than $2 billion in new funding to scale its drug discovery engine beyond protein structure prediction. The round, led by Thrive Capital with backing from Alphabet itself, reflects growing confidence that AI can accelerate one of the slowest, most expensive parts of medicine: finding and validating new drugs.

The funding announcement underscores a critical shift in how the pharmaceutical industry approaches research and development. Rather than relying solely on AlphaFold 3, the Nobel Prize-winning protein-folding model, Isomorphic Labs has built a more specialized tool called the Isomorphic Labs Drug Design Engine, or IsoDDE. This system tackles the messy, practical work that comes after scientists understand a protein's shape: finding where drugs can attach and predicting whether they will actually work.

What Makes IsoDDE Different From AlphaFold 3?

AlphaFold 3, released in 2024, delivered a more than 50% accuracy boost on certain research tasks and excels at analyzing small molecules, the chemical building blocks of most drugs. But IsoDDE goes further. When tested on a benchmark called Runs N' Poses, which measures how well a system can find binding pockets in proteins, IsoDDE more than doubled AlphaFold 3's score on the hardest tasks.

Binding pockets are the entry points through which drugs slip into disease-causing cells. Finding them manually takes researchers weeks or months. IsoDDE automates this process and requires far less preparatory data about a protein, reducing the upfront work needed before a drug project can begin.

Beyond pocket detection, IsoDDE predicts binding affinity, or how strongly a drug candidate will stick to its target protein, with significantly higher accuracy than earlier methods. This matters because a drug that binds too weakly will not work; one that binds too strongly may cause side effects.

How Is Isomorphic Labs Positioned in the Broader AI-Pharma Landscape?

Isomorphic Labs spun off from Google DeepMind in 2021 and is led by Demis Hassabis, who co-founded DeepMind and shared the 2024 Nobel Prize in Chemistry for AlphaFold 2 work. The company operates at the intersection of two powerful trends: the proven success of AI in decoding biology and the pharmaceutical industry's urgent need to speed up drug discovery.

Major pharma companies are already betting heavily on AI partnerships. Novartis, for example, has signed multibillion-dollar collaborations with Alphabet's Isomorphic Labs, Schrödinger, and Generate Biomedicines. Roche has built an "AI factory" with Nvidia GPUs, and Eli Lilly has partnered with the same. These moves reflect a recognition that AI is now central to pharmaceutical innovation, not a side experiment.

The appointment of Novartis CEO Dr. Vasant Narasimhan to the board of Anthropic, another major AI company, signals that pharma executives are taking active roles in shaping AI governance for healthcare. This cross-industry collaboration suggests that the pharmaceutical sector views AI governance as critical to responsible deployment of these tools in medicine.

What Will Isomorphic Labs Do With the $2 Billion?

According to the funding announcement, Isomorphic Labs plans to use the capital to enhance IsoDDE's capabilities and expand its international presence. The company aims to make its drug discovery engine more powerful and accessible to pharmaceutical partners worldwide.

This expansion comes as the AlphaFold Database, launched in 2021, has grown to contain structure predictions for over 200 million proteins, with more than 3 million users across 190 countries. The database continues to expand, with EMBL adding millions of AI-predicted protein complex structures through partnerships with Nvidia and Seoul National University.

Steps to Understand How AI Drug Discovery Works

• Protein Structure Prediction: AI models like AlphaFold 3 analyze amino acid sequences and predict how proteins fold into three-dimensional shapes, a task that once required years of laboratory work and X-ray crystallography.
• Binding Pocket Identification: IsoDDE scans the predicted protein structure to find binding pockets, the specific sites where drug molecules can attach and potentially neutralize disease-causing cells.
• Compound Screening and Affinity Prediction: The system predicts how strongly candidate drug compounds will bind to target proteins, filtering out weak binders and identifying the most promising candidates for further testing.
• Validation and Optimization: Researchers then validate top candidates in the laboratory and optimize their properties, a process that AI accelerates by reducing the number of failed experiments needed.

Why Does This Matter for Drug Development Timelines?

Traditional drug discovery takes 10 to 15 years and costs billions of dollars. The bottleneck is not just understanding disease biology; it is finding molecules that work safely and effectively. AI tools like IsoDDE compress the early stages of this process by automating tasks that once required teams of chemists and structural biologists working in parallel.

The practical impact is significant. By reducing the amount of preparatory data needed and automating binding pocket detection, IsoDDE lowers the barrier to starting new drug projects. Researchers can move from identifying a disease target to testing candidate compounds faster, potentially bringing new medicines to patients years earlier.

The $2 billion funding round reflects investor confidence that this acceleration is real and scalable. Thrive Capital, which backs OpenAI, sees enough promise in Isomorphic Labs to lead the round alongside Alphabet. This signals that AI-driven drug discovery is moving from research labs into commercial deployment.

"The solution the Google team discovered using AlphaEvolve unlocks meaningfully higher accuracy rates for our sequencing instruments. For researchers, this higher-quality data might enable the discovery of previously hidden disease-causing mutations," stated Aaron Wenger, Senior Director at PacBio.

Aaron Wenger, Senior Director at PacBio

Beyond Isomorphic Labs, the broader ecosystem of AI-powered drug discovery is expanding. Google DeepMind's AlphaEvolve, a Gemini-powered coding agent, has been applied to improve DeepConsensus, a model for correcting DNA sequencing errors, achieving a 30% reduction in variant detection errors. In computational materials and life sciences, Schrödinger used AlphaEvolve to achieve roughly a 4x speedup in both Machine Learned Force Fields training and inference, directly shortening R&D cycles in drug discovery.

The convergence of these advances suggests that AI is not replacing biologists and chemists; rather, it is amplifying their capabilities. Researchers can now explore larger chemical spaces faster, test more hypotheses in silico before moving to the lab, and focus their expertise on the creative and strategic aspects of drug discovery rather than routine computational tasks.

As Isomorphic Labs scales with its new funding, the pharmaceutical industry will likely accelerate its shift toward AI-native drug discovery workflows. The question is no longer whether AI can help find new drugs, but how quickly companies can integrate these tools into their existing R&D pipelines and train their teams to work alongside AI systems.