The Dawn of Self-Driving Labs: How AI and Robots Reinvent Medicine

SALT LAKE CITY, UT – February 23, 2026 – The painstaking, years-long process of discovering new medicines may soon be a relic of the past. A new frontier is opening, not in a remote jungle or deep sea trench, but inside highly automated laboratories where artificial intelligence and robotics are taking the lead. This transformation will be in the spotlight at the upcoming NVIDIA GTC AI Conference, where clinical-stage TechBio company Recursion will be a key feature in a talk on the future of “self-driving labs.”

The presentation, hosted by laboratory automation leader HighRes Biosolutions, is titled “AI Agents, Robotics, and Digital Twins: The Full Stack of Self-Driving Labs and Biomanufacturing.” It signals a pivotal moment for the pharmaceutical industry, showcasing a collaboration that aims to automate and accelerate drug discovery on an unprecedented scale. This isn't just about making labs more efficient; it's about fundamentally redesigning the engine of scientific discovery itself.

The Rise of the Automated Scientist

At the heart of this revolution is the concept of the “self-driving lab,” a closed-loop system where AI designs experiments, robots execute them, and machine learning algorithms analyze the results to inform the next cycle of discovery—all with minimal human intervention. This vision is being made possible by a convergence of powerful technologies.

HighRes Biosolutions, a key partner in this ecosystem, provides the foundational building blocks. CEO Ira Hoffman, who will lead the GTC presentation, is steering the company's efforts in three critical areas:

• Robotic Perception: This technology gives robots the ability to “see” and interact with their environment with high precision, moving beyond simple, repetitive tasks to perform complex experimental procedures adaptively.
• Digital Twins: HighRes creates highly detailed virtual replicas of physical laboratories. These digital twins, often built on platforms like NVIDIA’s Omniverse, allow scientists to simulate, test, and optimize entire workflows in a virtual space before committing a single physical resource. This dramatically reduces errors and accelerates setup.
• Natural Language Orchestration: Perhaps the most transformative element, this allows scientists to direct complex robotic systems using simple, human-like commands, breaking down the barrier between biologist and machine.

Together, these technologies form the backbone of a lab that can run 24/7, tirelessly exploring millions of biological and chemical interactions far beyond the scope of human capacity.

Recursion’s High-Stakes Bet on 'Physical AI'

For Salt Lake City-based Recursion, this collaboration is the culmination of more than a decade of strategic investment in building one of the world's most advanced drug discovery platforms. The company has long understood that the future of biology is deeply intertwined with data and computation. Its integrated wet and dry labs already generate millions of multi-omic data points each week, feeding a proprietary dataset of biological and chemical information that is staggering in its scale.

This torrent of data is the lifeblood for training Recursion’s sophisticated machine learning models. However, processing it requires immense computational power. This is where the company’s NVIDIA-backed supercomputer, BioHive-2, becomes a critical competitive advantage. By owning and operating one of the most powerful supercomputers on the planet, Recursion can distill trillions of searchable relationships from its data, uncovering novel biological insights that are unconstrained by human bias.

This fusion of massive-scale robotics and massive-scale computing is what the industry is beginning to call “Physical AI.” It represents a move to bring artificial intelligence out of the digital realm and into the physical world to perform meaningful work. As stated by Rory Kelleher, NVIDIA's Global Head of Business Development for Healthcare, “This year at GTC, we’re collaborating across the ecosystem to move Physical AI from concept to reality, unlocking a new era of biological insight and therapeutic speed.”

For Recursion, this isn't just a research project; it's a core business strategy. In a competitive landscape populated by other AI drug discovery firms like BenevolentAI and Exscientia, Recursion's deep investment in physical infrastructure creates a significant moat. While others focus primarily on software and data analysis, Recursion is building an end-to-end system that controls the entire discovery process, from initial hypothesis to clinical application.

A New Frontier for Medicine and Manpower

The ultimate promise of this technological leap is its potential human impact. By dramatically accelerating the early stages of drug discovery, self-driving labs could shorten the timeline for bringing new therapies to patients suffering from a wide range of diseases. The ability to screen vast chemical libraries against complex biological models could unlock treatments for conditions long considered undruggable.

However, the rise of the automated scientist also raises questions about the future of the scientific workforce. The vision is not one of replacement, but of augmentation. By automating the laborious and repetitive tasks of pipetting, sample preparation, and data collection, these systems free up human scientists to focus on what they do best: strategic thinking, creative problem-solving, and formulating the next big hypothesis. The goal is to elevate the role of the researcher from a lab technician to a campaign strategist, directing armies of robotic assistants in the search for cures.

Significant challenges remain. Integrating a multitude of complex robotic, software, and analytical systems is a monumental engineering task. Ensuring the quality and reproducibility of data generated at such a massive scale requires rigorous validation. Furthermore, gaining regulatory acceptance for drugs discovered through largely autonomous processes will require building new frameworks for trust and transparency. These are the hurdles that companies like Recursion and HighRes are now working to overcome.

The GTC Spotlight: A Convergence of Titans

The upcoming lightning talk at NVIDIA GTC is more than just a corporate presentation; it’s a snapshot of a burgeoning industry-wide movement. The session, moderated by NVIDIA’s Stacie Calad-Thomson, a veteran of GSK and the Alliance for Artificial Intelligence in Healthcare, brings together a diverse group of leaders. Alongside HighRes’s Ira Hoffman will be Fred Parietti, CEO of robotics biomanufacturing firm Multiply Labs, and Olga Ovchinnikova of Thermo Fisher Scientific, highlighting the broad ecosystem of players required to realize this future.

This convergence of expertise—from supercomputing and AI to robotics, automation, and deep biological science—underscores the interdisciplinary nature of modern R&D. The session in San Jose will offer a compelling glimpse into a future where the speed of biological discovery is no longer limited by human hands, but only by the power of our imagination and the machines we build to realize it.