Lila Sciences just entered the elite circle of AI-first biotech companies by crossing a $1.3 billion valuation. The San Francisco–based startup raised an additional $115 million this week, extending its Series A round to $350 million. Nvidia’s venture capital arm led this new funding, with strong participation from Sequoia Capital, Andreessen Horowitz, and Lux Capital.
This funding marks a turning point not only for Lila but also for the broader AI-powered scientific research sector. Lila’s team doesn’t just train large language models; they build “AI Science Factories” — automated laboratories where robotic systems, sensors, and deep learning models collaborate to accelerate scientific discovery.
Building the “AI Science Factory”
Lila Sciences began in 2022 when Dr. Sonia Kim, an AI researcher from MIT, joined forces with molecular biologist Dr. Aaron Velasquez. Both shared one frustration — human-led experiments couldn’t keep up with the speed of AI hypothesis generation. AI could predict chemical reactions or protein behaviors in seconds, but the physical verification of those predictions still took weeks or months.
They decided to close that gap.
Lila’s first major invention, LilaLab OS, merged real-time robotics, high-throughput data collection, and reinforcement learning. The system runs continuous experiments, measures results, feeds data back into AI models, and instantly adjusts future experiments — all without human intervention.
Dr. Kim described the process during a recent press conference:
“We built an environment where AI doesn’t stop at thinking — it actually acts. The lab becomes an intelligent organism, capable of learning through doing.”
This setup allows Lila to run thousands of biochemical experiments every day. Each experiment improves its models’ understanding of molecular interactions, reaction efficiencies, and synthesis routes. Essentially, the AI learns science through trial and error, just like a human scientist — but millions of times faster.
Nvidia’s Strategic Investment
Nvidia viewed Lila’s work as a natural extension of its AI computing ecosystem. The company already dominates AI model training with its GPUs, but it wants to push into AI-driven physical automation.
Nvidia’s venture division led the $115 million extension round, and Jensen Huang personally met Lila’s founders before finalizing the deal. In a joint statement, Nvidia explained that Lila’s combination of robotics, sensors, and AI directly aligns with its push toward AI-accelerated scientific simulation.
Nvidia plans to supply Lila with its latest DGX H200 clusters and Omniverse simulation tools. These will help Lila’s systems visualize molecular interactions in 3D, simulate new materials, and design optimized experimental sequences before physical execution.
Dr. Velasquez praised the partnership:
“Nvidia gives us computational muscle and simulation capability. With that power, our labs can design, test, and refine experiments almost in real time.”
Lila’s leadership believes this partnership transforms the startup from a fast-moving AI company into a deep-tech powerhouse with unmatched experimental throughput.
From AI Models to Scientific Breakthroughs
Lila’s early successes already demonstrate the value of its approach.
The company focused first on enzyme optimization, a field essential for biofuel production and pharmaceutical synthesis. Traditional labs need months to find mutations that improve enzyme performance. Lila’s platform completed that process in under two weeks.
Its AI models generated thousands of enzyme variants, and robotic systems synthesized and tested them autonomously. The system then retrained itself on the new data, narrowing in on the best-performing candidates.
Lila didn’t stop there. The team expanded into materials discovery, seeking new polymers that degrade faster in the environment. Within three months, their system identified five novel polymer structures that showed 80 % faster biodegradation than current materials.
Every success added credibility to Lila’s core thesis: automation plus AI can compress the entire cycle of scientific innovation.
A New Category in Science
Lila defines its mission as creating “self-improving scientific systems.” The company no longer hires traditional research scientists in large numbers; instead, it hires AI engineers, robotics specialists, and systems designers who teach machines how to do science.
Dr. Kim often compares Lila’s vision to building “a brain inside a lab.” The system senses the world, thinks about what it perceives, and takes physical actions to learn more.
She explained the philosophy during her keynote at the GITEX 2025 tech summit:
“Human intuition remains irreplaceable, but humans don’t need to pipette liquids or analyze spectra manually. AI can handle that grind. Scientists should focus on setting goals and interpreting meaning.”
This approach has already started influencing how major research institutions operate. Several universities and pharmaceutical companies licensed LilaLab OS to automate parts of their experimentation pipelines.
The shift points toward a near future where AI doesn’t just support scientific discovery — it performs it.
Expanding into Global Science Hubs
After its Series A extension, Lila plans to expand operations beyond the United States. The company will open new AI Science Factories in Cambridge (UK), Singapore, and Bengaluru (India).
The Bengaluru center will focus on AI-driven drug discovery, collaborating with Indian pharmaceutical startups and university researchers. Lila chose the region for its deep talent pool in both biotech and AI engineering.
Meanwhile, the Singapore facility will specialize in synthetic biology and materials science, while the Cambridge site will emphasize AI ethics and scientific reproducibility.
Each hub will run independently but connect through a shared data infrastructure. Every experiment conducted anywhere in the world will feed into Lila’s central AI knowledge base, continuously improving the entire system.
Dr. Velasquez summarized this strategy:
“We don’t want isolated labs. We want one global scientific network where every robot learns from every other robot.”
Competitive Landscape
Lila now stands at the forefront of a new competitive field. Startups like Emergent BioAI, DeepScience Labs, and Atomwise also combine AI and experimentation, but none integrate automation as deeply.
Emergent BioAI still relies on human technicians for experiment setup. DeepScience Labs focuses on simulation rather than physical experiments. Atomwise specializes in AI-based drug screening but lacks its own robotic infrastructure.
Lila combines all three components — simulation, automation, and experimentation — into one continuous loop. This integration creates a strong moat that competitors may struggle to replicate quickly.
Industry analysts now call Lila’s model “closed-loop science.” The system generates hypotheses, executes experiments, and learns from outcomes without manual steps in between.
Impact on Scientific Research and Industry
Lila’s breakthrough matters far beyond its valuation. The company demonstrates that science no longer needs to follow the slow, linear path of hypothesis, testing, and review. AI systems can compress those steps into tight feedback loops.
Pharmaceutical companies see enormous potential. Lila already signed early partnerships with two unnamed global drug makers to speed up compound screening. In agriculture, the firm collaborates with AgriNext to develop AI-optimized enzymes for sustainable fertilizers.
Environmental researchers also benefit from Lila’s biodegradable materials program, which could revolutionize packaging and consumer products.
Even government agencies have taken notice. The U.S. Department of Energy reportedly approached Lila to explore automated material discovery for clean-energy applications.
Dr. Kim believes this is only the beginning:
“We want to make discovery continuous. Instead of waiting for breakthroughs every few years, we can create a world where science evolves daily.”
A Future Driven by Intelligent Laboratories
The story of Lila Sciences shows what happens when artificial intelligence stops being a tool and becomes a collaborator. The company’s fusion of robotics, deep learning, and experimental feedback redefines how humanity approaches discovery.
Every day, its labs run thousands of autonomous experiments, and each one pushes the boundary of what machines can understand about nature. Nvidia’s investment validates this direction, signaling that the intersection of AI and physical experimentation will define the next era of innovation.
Lila doesn’t just chase a higher valuation. It builds a new framework for how the world can pursue truth faster, cheaper, and smarter. The company’s founders envision a decade when AI systems in labs across continents collaborate like neurons in a global scientific brain.
If that vision holds true, Lila Sciences may not only represent a new kind of startup — it may represent a new kind of science altogether.
Also Read – Menstrual Leave Policy at the Workplace: Why It’s Important