Google's secret weapon against cancer might be an AI algorithm

Alphabet’s AI-powered drug discovery company, Isomorphic Labs, is preparing to take its first major leap into the real world. After years of development, the company says it’s now gearing up for human clinical trials of cancer drugs, treatments designed entirely by artificial intelligence.

The company was spun out of Google DeepMind in 2021, and is built on the breakthrough AlphaFold system, an AI model that stunned the scientific community by accurately predicting the 3D structure of proteins, a fundamental building block of biology. That innovation has since evolved into AlphaFold 3, capable of modelling how proteins interact with DNA, RNA, and potential drug molecules, effectively allowing researchers to simulate parts of the drug development process before a single test tube is touched.

This shift could mean a faster, cheaper, and more targeted way to develop medicine, a bold reimagining of a process that typically takes over a decade and billions of dollars to bring a single drug to market.

In a recent interview with Fortune, Isomorphic Labs President Colin Murdoch said the company is “very close” to starting clinical trials. “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,” he said.

The company’s approach combines its AI expertise with traditional pharmaceutical know-how. Beyond its internal drug programmes, Isomorphic is also working with pharmaceutical giants like Novartis and Eli Lilly, both of which signed multi-year research partnerships with the company in 2024. In April this year, Isomorphic raised $600 million in its first external funding round, led by Thrive Capital. That funding is now being channelled into advancing in-house drug candidates, with oncology and immunology among the early focus areas.

Murdoch says the aim is to build a “world-class drug design engine,” one that could eventually shift the odds in favour of drugmakers, who currently face high failure rates and massive R&D costs. “We’re trying to speed them up, reduce the cost, but also really improve the chance that we can be successful,” he said.

But the use of AI in healthcare, particularly at such a critical stage of development, raises as many questions as it does possibilities. While the promise of faster cures is enticing, concerns around transparency, access, and accountability are already emerging.

The so-called “black box” nature of AI, where the system delivers results without a clear understanding of how it got there, could prove especially challenging in medicine, where trust, reliability, and regulatory compliance are non-negotiable. There are also broader questions: Who owns the rights to a drug created by AI? Will treatments be affordable and accessible, or locked behind patents and licensing deals? And if something goes wrong, who’s responsible: the developers, the AI model, or the company?

When contacted by Gizmodo, Isomorphic Labs said it had “nothing more to share” at this time.

What’s clear is that this marks a significant turning point, not just for Alphabet, but for the pharmaceutical industry as a whole. If Isomorphic’s trials succeed, it could reshape how we discover and deliver medicine. But that future will depend not only on what the AI can do, but on whether regulators, ethicists, and the public are ready to accept it.