Edison Scientific Raises $70M to build autonomous AI scientists Platform Kosmos
San Francisco–based Edison Scientific has emerged from stealth with a $70 million seed round, an unusually large raise for a company at this stage and a clear signal of investor confidence in AI-enabled scientific research. The round was co-led by Spark Capital and Triatomic Capital, with participation from a major U.S. institutional biotech investor, as well as existing backers Pillar VC and Susa Ventures. Additional support comes from Striker Venture Partners, Hawktail VC, Olive VC, and a network of prominent angels, including leaders from top AI labs and major pharmaceutical companies.
This funding will support the development and deployment of Edison Scientific’s AI platform Kosmos, which is designed to automate significant portions of the research process across biology, chemistry, and materials science.
From FutureHouse Spin-Out to Stand-Alone AI Science Company
Edison Scientific was spun out of FutureHouse, a research and venture creation organization focused on accelerating progress in science through AI and new institutional models. By building on infrastructure, expertise, and community developed within FutureHouse, Edison Scientific begins life with a mature network of collaborators and early adopters rather than as a traditional early-stage startup.
The spin-out reflects a broader trend in deep-tech entrepreneurship, where venture creation engines incubate highly technical companies before launching them with substantial capital and clearly defined product roadmaps. Edison Scientific intends to operate as a bridge between AI advances and practical laboratory science, seeking to reduce the time between hypothesis and validated results.
Kosmos: An AI Scientist Designed to Run Hundreds of Research Tasks in Parallel
At the core of Edison Scientific’s offering is Kosmos, an “AI scientist” intended to augment human researchers by automating key elements of the scientific workflow. Rather than focusing solely on language generation or code writing, Kosmos is built to orchestrate hundreds of research tasks simultaneously, from literature synthesis and study interpretation to data analysis and molecular design.
According to the company, Kosmos can transform raw datasets into structured, validated reports, compressing processes that might normally take researchers months into a single computational run. Human scientists remain in control, but the platform aims to remove repetitive bottlenecks that slow discovery across academia and industry.
Automating Literature Review, Data Analysis, and Molecular Design
One of the major challenges in modern science is not the lack of data, but the overwhelming volume of it. Researchers face exponentially growing literature, fragmented datasets, and complex experimental outputs that must be interpreted before new work can even begin. Edison Scientific’s platform is designed to synthesize existing research, analyze large datasets and propose next steps in experimental design.
In areas such as drug discovery, protein engineering, or materials development, the ability to iterate faster on candidate designs could significantly shorten R&D cycles. The promise is not just speed, but the ability to systematically explore ideas that human teams might overlook due to time or resource constraints.
Potential Impact Across Biotech, Pharma, and Materials Science
If Edison Scientific’s technology delivers as promised, its implications could extend across multiple scientific domains. In biopharma, AI-accelerated discovery could shorten the timeline from target identification to preclinical validation. In materials science, Kosmos-assisted design processes could enable faster development of batteries, catalysts, or industrial compounds.
Even in academic research environments, AI-assisted synthesis of literature and results could free scientists to spend less time compiling reviews and more time designing experiments. The company emphasizes that AI does not replace scientific expertise; instead, it serves as an amplifier of human capability, allowing smaller teams to pursue more ambitious research agendas.
Leadership With Deep Roots in Science and Technology
Edison Scientific is led by Sam Rodriques, physicist, bioengineer, and co-founder of both FutureHouse and Edison Scientific. Rodriques previously conducted research at the Francis Crick Institute and has been involved in the invention of technologies spanning transcriptomics, brain mapping, nanofabrication, and gene therapy.
His scientific background, combined with Edison Scientific’s bench of AI engineers and research partners, is intended to build credibility with the scientific community, an essential factor in an era where claims about “AI scientists” can attract skepticism.
By pairing domain expertise with compute infrastructure and venture backing, Edison Scientific aims to position itself as a serious scientific partner focused on real-world discovery outcomes.
Claims of “AI scientists” are easy to make and hard to substantiate, but Edison Scientific’s combination of experienced leadership, deep scientific grounding, and substantial early funding makes it worth watching closely. If Kosmos meaningfully reduces the time between question and experimentally validated answer, it could reshape the economics of research and change how labs, startups, and major pharma companies approach discovery itself.