Helical raises $10m to turn powerful AI models into practical systems that pharma teams can use and trust in real-world drug discovery.
If we zoom it out, modern medicine looks incredibly productive. Billions poured into research, breakthroughs announced every week, AI models getting smarter by the day. However, it’s only necessary that we face the truth that most ideas never make it past the testing phase.
Roughly 50 new drugs are approved each year, despite more than 10,000 known diseases. It’s not because scientists aren’t thinking hard enough. It’s because every idea still has to survive the same slow, expensive and failure-prone physical testing process.
Before anything reaches a patient, it has to pass through years of lab work, trials and validation. And more than 90% of drug candidates fail along the way.
That’s the bottleneck Helical is trying to break. The London-based AI startup has raised $10 million in seed funding, led by redalpine with participation from Gradient, BoxGroup, Frst and a group of notable angels, to build what it calls a “virtual AI lab” for drug discovery [1].
AI can predict, but can it decide?
There’s been no shortage of excitement around AI in pharma, especially with the rise of biological “foundation models,” systems trained on massive datasets of genes, proteins and clinical information.
In theory, these models let scientists test ideas on a computer before stepping into a lab. Imagine being able to simulate thousands of experiments in the time it takes to run one in real life. In practice, though, things tend to stall.
Teams generate predictions, but those predictions often live in isolated notebooks, scattered across departments or locked behind technical workflows that only a handful of specialists understand. Biologists and machine learning engineers speak different languages, work toward the same goal, but rarely within the same system.
What you get is insight without momentum. Helical’s core idea is that the problem isn’t the models; it’s everything around them.
“The models alone don’t discover drugs. The system does,” said Rick Schneider, co-founder of Helical. “Pharma teams need a system that turns foundation models into workflows scientists can run, validate, and defend. We built Helical to make in-silico science reproducible at pharma scale, so teams can go from hypothesis to decision in days instead of months.”
A lab you don’t have to step into
Helical’s platform serves as a shared workspace where experiments occur digitally before they do physically.
The company splits it into two parts: a “Virtual Lab” for biologists and translational scientists, and a “Model Factory” for machine learning engineers and data scientists. But both operate on the same data, the same models and crucially the same results.
Instead of one-off experiments that live and die on someone’s laptop, workflows become repeatable. Results can be checked. Decisions can be traced back to evidence. In science, that’s everything, because a fast answer that can’t be trusted is useless and a promising result that can’t be reproduced is a dead end.
Helical’s timing isn’t accidental. Pharma R&D spending now exceeds $300 billion annually, with timelines stretching beyond a decade and average costs surpassing $2 billion per drug. For all that investment, the output remains stubbornly limited.
AI was supposed to bend that curve. However, many efforts have stayed stuck in pilot mode, impressive demos that don’t quite translate into real-world decisions. Investors are starting to recognize the gap.
“We are at a unique point in time where biological foundation models and general language reasoning models are converging,” said Daniel Graf, General Partner at redalpine. “We backed Helical because we strongly believe they have what it takes to build the pharma AI orchestration platform that will drive this transition from siloed AI models to integrated virtual AI labs.”
The phrase “orchestration platform” might sound abstract, but the idea is to simply connect the pieces so the whole system works.
Early signals from the field
Helical is still young, founded in early 2024 by three school friends who later branched into tech, data science and medicine, but it’s already working with multiple top-20 global pharma companies. That includes a public collaboration with Pfizer on predictive blood-based safety biomarkers, an area where early detection can save years of wasted development.
Across deployments in target identification, biomarker discovery and therapeutic design, the company says teams have compressed timelines from years to weeks. Bold, yes, but even partial progress toward that goal could reshape how drugs are developed.
Longevity research stands to gain
For those working in longevity, the implications run deeper. Aging isn’t a single disease with a single target. It’s a network of processes – cellular damage, inflammation, metabolic shifts – interacting in ways we’re still trying to map.
Traditional drug discovery struggles with that kind of complexity. It’s too slow, too linear, too dependent on testing one idea at a time. A system that allows scientists to explore multiple hypotheses simultaneously and quickly discard those that don’t work could change the equation. Not by replacing the lab, but by making every lab experiment more intentional.
If Helical’s approach holds up, it could mean fewer wasted years chasing dead ends and more focus on the mechanisms that actually move the needle on aging.
Helical is not alone in trying to industrialize AI for drug discovery across longevity-linked biology. In the broader longevity biotech landscape, several better-capitalized players are already pushing variations of that thesis at scale. According to DLT, these include Recursion, which has raised about $1.62 billion, Insilico Medicine with well over $1 billion raised and a deal with Eli Lilly earlier this year valued at up to $2.75 billion, BioAge Labs at around $648 million and Enveda at close to $530 million.
Even newer longevity-heavy entrants such as Retro Biosciences, with about $1.18 billion raised, are part of the same wider shift toward computationally guided biology. What stands out is that these are not tiny science projects: Insilico alone shows 22 assets in the tracked pipeline, while Recursion has 7, BioAge 5 and Enveda 4. That matters because the AI drug discovery story in longevity is starting to look less like a speculative side bet and more like an emerging competitive layer, one where the real question is no longer whether algorithms can generate hypotheses, but which companies can turn those hypotheses into repeatable programs, credible evidence and eventually approved medicines.
Helical plans to use its new funding to expand into more therapeutic areas, deepen collaborations with existing pharma partners and build what it calls a “compounding evidence layer,” a system that improves as more experiments, data and decisions pass through it.
For years, the conversation around AI in drug discovery has been about potential. Helical is changing that, claiming that the next phase is about what systems can reliably deliver.
Photograph courtesy of Helical
