Skip to the main content.

4 min read

AI-Designed Drugs Moving to Human Trials

AI-Designed Drugs Moving to Human Trials
AI-Designed Drugs Moving to Human Trials
8:16

We're living through the opening scene of what might be the most consequential medical thriller of our time. Google DeepMind's Isomorphic Labs is preparing to inject the first AI-designed drugs into human veins by the end of 2025—a moment that feels simultaneously inevitable and utterly surreal. Think about it: molecules conceived not in test tubes but in silicon minds, crafted by algorithms that have digested more protein structures than a thousand doctoral students could process in their lifetimes.

This isn't science fiction posing as press release. This is the logical culmination of a journey that began when DeepMind's AlphaFold cracked the protein folding problem—a feat that earned Demis Hassabis and John Jumper the 2024 Nobel Prize in Chemistry. But here's what makes us lean forward rather than back: the brutal honesty of the numbers driving this revolution.

The Beautiful Mathematics of Desperation

Let's talk about the pharmaceutical industry's dirty little secret: an average likelihood of first approval rate of 14.3% across leading research-based pharmaceutical companies, with some companies struggling with success rates as low as 8%. The traditional drug discovery pipeline is a $2.6 billion, 12-to-18-year exercise in institutionalized hope—a process so inefficient it would make a Soviet factory planner blush.

On average, it can take anywhere between 10 and 15 years — and sometimes more than $2 billion — for a drug to make it from the bench to final approval from the Food and Drug Administration. Meanwhile, the global drug discovery market is projected to explode from $110.26 billion in 2024 to $125.23 billion in 2025, driven by the kind of desperation that breeds innovation.

Enter Colin Murdoch, Isomorphic Labs' president, who recently told Fortune something that should make every pharmaceutical executive either very excited or very nervous: "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."

This isn't algorithmic theater. In April 2025, Isomorphic Labs raised $600 million in its first-ever external funding round, led by Thrive Capital. Google doesn't write nine-figure checks for science projects. They're betting on a future where Murdoch's vision becomes reality: "One day we hope to be able to say — well, here's a disease, and then click a button and out pops the design for a drug to address that disease."

New call-to-action

The Zeitgeist of Silicon-Powered Medicine

What makes this moment particularly fascinating is how it mirrors our broader cultural obsession with artificial intelligence as both savior and potential destroyer. We're the generation that grew up with Terminator movies and now asks ChatGPT to write our emails. We're simultaneously terrified of AI and utterly dependent on it for everything from navigating traffic to choosing what to binge-watch.

But here's where pharmaceutical AI differs from the flashy consumer applications: the stakes are existential, and the timeline is measured in decades, not app store updates. By 2025, it is estimated that 30% of new drugs will be discovered using AI, and AI has been shown to reduce drug discovery timelines and costs by 25-50% in preclinical stages.

The industry is throwing money at this problem with the urgency of a drowning man reaching for a life preserver. Spending on AI will reach $3 billion by 2025 as companies invest in technology that may reduce the time and costs required to bring a new drug to market. AI-based drug discovery alliances are also increasing, from just 10 in 2015 to 105 in 2021.

The Prometheus Moment

Here's where our cautious optimism kicks in. Unlike the social media algorithms that have spent the last decade optimizing for engagement and accidentally optimizing for rage, pharmaceutical AI is optimizing for something profoundly more important: keeping humans alive. The incentive structures are aligned in ways that make Silicon Valley's "move fast and break things" ethos suddenly feel mature and measured.

Isomorphic Labs isn't just throwing neural networks at the problem and hoping for the best. They're building on AlphaFold's proven foundation—a system that predicted the structure of virtually all the 200 million proteins that researchers have identified, something that would have otherwise taken a billion years of PhD time to complete. This isn't speculative technology; it's battle-tested science that has already proven its worth in laboratories worldwide.

The partnerships tell the story: In 2024, the same year it released AlphaFold 3, Isomorphic signed major research collaborations with pharma companies Novartis and Eli Lilly. These aren't companies known for betting on vaporware. They're placing strategic bets on a future where AI doesn't just assist drug discovery—it leads it.

The Moonshot That's Already Leaving the Launchpad

What makes this moment feel like a genuine inflection point rather than another overhyped tech demo is the convergence of several trends. The global drug discovery market size was valued at USD 24.84 billion in 2024 and is projected to reach from USD 28.98 billion in 2025 to USD 62.10 billion by 2033, growing at a CAGR of 14.90%. This isn't growth—it's a moonshot disguised as a market trend.

Meanwhile, the global artificial intelligence in drug discovery market size was estimated at USD 1.5 billion in 2023, and is expected to expand at a CAGR of 29.7% from 2024 to 2030. The math is clear: AI is eating pharmaceutical R&D from the inside out, and the early adopters are positioning themselves to feast.

The human element remains crucial. As Hassabis noted, "True invention is not possible yet with AI. It can't come up with a new hypothesis or new conjecture". But what AI can do—and what Isomorphic Labs is betting billions on—is execute the grinding, computational heavy lifting that has historically consumed years of human genius.

The Careful Revolution

Our cautious optimism stems from recognizing that this isn't a disruption in the classic Silicon Valley sense. It's more like the introduction of antibiotics or the discovery of DNA—a fundamental shift in capability that will ripple through medicine for generations. The timeline isn't measured in quarters but in decades, and the stakes aren't user engagement but human survival.

Hassabis said that AI could eventually help to bring down time to get from target to a candidate to months or weeks, and he envisions "personalised medicine where it's optimised maybe overnight by an AI system for your personal metabolism to be perfect for you". It's a future that sounds like science fiction but feels increasingly like science fact.

The pharmaceutical industry has always been a game of massive risks and massive rewards. What's changing is the mathematics of those risks. AI isn't eliminating uncertainty—it's making uncertainty more intelligent, more targeted, and potentially more successful.

As we stand on the brink of human trials for AI-designed drugs, we're witnessing the birth of a new kind of medicine—one where silicon minds and human hearts collaborate to solve the oldest problems in the newest ways. The future isn't just arriving; it's asking for informed consent.


Ready to harness AI's transformative power for your marketing and growth strategy? At Winsome Marketing, our growth experts help companies navigate the intersection of artificial intelligence and business success. Let's discuss how AI can revolutionize your approach to customer acquisition, retention, and revenue growth. Contact our team today to explore your AI-powered future.

AI That Thinks Like a Clinician

AI That Thinks Like a Clinician

Doctors don't just work during work hours—they work in their pajamas. After long shifts caring for patients, physicians spend another 86 minutes...

READ THIS ESSAY
Microsoft's AI

Microsoft's AI "Outperforms" Doctors

Microsoft just announced their AI system can diagnose complex medical cases with 80% accuracy while human doctors managed only 20% on the same test...

READ THIS ESSAY
Renovaro's New AI Drug Discovery Patent

4 min read

Renovaro's New AI Drug Discovery Patent

In an era when most AI patents feel like elaborate ways to rebrand existing technology, Renovaro's latest USPTO approval stands out for all the...

READ THIS ESSAY