Autoscience Secures New Funding to Scale Its Autonomous AI Research Lab
Autoscience, an applied research lab based in San Mateo, CA, has raised $14M in seed funding to automate the research and development of new machine learning models. The funds will be used to advance its virtual AI laboratory made of non-human AI scientists and engineers. The startup claims that its lab can “invent, validate, and deploy specialized, state-of-the-art machine learning models.”
This funding round was led by General Catalyst and also included Toyota Ventures, Perplexity Fund, S32, and MaC Ventures. The latest funding round comes only one year after the startup published a full-length research paper generated end-to-end by its AI agent Carl that successfully passed peer review.
The virtual AI lab was developed to overcome what it claims is the primary bottleneck in AI development – the human capacity to create and test new ideas at scale. By letting AI systems handle more of the research work, the company hopes to speed up how quickly new models can be built and tested.
(Rawpixel.com/Shutterstock)
The idea reflects a growing belief that parts of the AI development process itself can be automated.
“We’ve reached a point where human intuition is no longer enough to navigate the complexity of algorithmic discovery,” said Eliot Cowan, CEO of Autoscience. “We’ve built a research organization where the researchers are AI systems. We aim to compress a decade of machine learning research into months, unlocking new AI capabilities for scientists and forming a competitive edge for our customers.”
The idea behind the virtual lab comes as machine learning research keeps getting harder to manage with human teams alone. Modern models require large numbers of experiments, constant testing, and a lot of trial and error.
Companies are starting to look at AI systems as a way to handle some of that work, so researchers can move faster and try more ideas than would normally be possible.
“We believe Autoscience is tackling an increasingly important challenge in machine learning: the pace and scalability of experimentation,” said Yuri Sagalov, Managing Director at General Catalyst. “As research output continues to grow, teams are looking for ways to more efficiently test, validate, and translate new ideas into production systems. We’re excited about their progress in advancing autonomous R&D to scale that workflow.”
The idea of AI systems helping build new AI models has also been raised by leaders across the industry. Speaking at the World Economic Forum in Davos earlier this year, Anthropic CEO Dario Amodei said one of the most important developments to watch is the growing role of automation inside AI research itself.
“I think the biggest thing to watch is this issue of AI systems building AI systems,” Amodei said, pointing to a future where more of the development process is handled by machines rather than human researchers.
(Sergey Nivens/Shutterstock)
The push toward automated research is also showing up in scientific computing and AI for science efforts. National labs, universities, and large tech companies are building systems that can run simulations, generate code, and search for new models using AI, with the goal of speeding up discovery in fields such as materials science, biology, and climate research.
One of the concerns around automated research is reliability. Systems that generate models or run experiments on their own may produce results that look correct but fail under real-world conditions. Ensuring that automatically generated models are accurate, safe, and reproducible remains an open problem, especially as workflows become more complex.
Another challenge is the amount of compute required to support this kind of automated experimentation. Running large numbers of model searches and training cycles can quickly become expensive, meaning that only well-funded labs and companies may be able to use these systems at scale.
Having said that, breakthroughs like Autoscience highlight the potential for AI in the field of science and research. Not as a replacement for human judgment and experience, but as a tool that can help researchers test more ideas, work faster, and explore problems that could otherwise take years to solve.
Related
OpenAI Shutters Sora, Shifts Business Strategy Ahead of IPO
Back in 2022, OpenAI set off a chain reaction in the tech world when it…
LANL Develops Diffusion AI Model for Electroplating Process Optimization
March 26, 2026 — Electrochemical deposition, or electroplating, is a common industrial technique that coats materials…
PNNL: AI Effort Moves from Novelty to Front Lines of National Lab’s Cyber Protection
RICHLAND, Wash., March 26, 2026 — A research effort to explore how artificial intelligence can offer…
ORNL Introduces ‘Photon’ Framework for Accelerating AI Vulnerability Discovery on Frontier
March 25, 2026 — Oak Ridge National Laboratory’s Center for Artificial Intelligence Security Research (CAISER) is…
Gartner Forecasts 90% Drop in LLM Inference Costs by 2030
STAMFORD, Conn., March 25, 2026 — By 2030, performing inference on a large language model (LLM)…
Dario Gil at GTC: DOE’s Genesis Mission Moving from Vision to Execution
At Nvidia’s GTC conference last week, a fireside chat between Nvidia’s Ian Buck and U.S. Department of Energy Under…
