Potato's AI research assistant searches publications for inspiration for scientific experiments


Potato co-founders Nick Edwards, left, and Ryan Kosai. (Potato photo)

Science can reward its practitioners with exciting eureka moments, with an emphasis on “moments.” Much of the research is done through experiments that don't always work, reviewing research publications, and engaging in mental contortions in search of the next hypothesis.

A new startup in the Seattle area aims to change that. Co-founders Ryan Kosai and Nick Edwards have built potato AI that they are pitching as AI scientific research assistant.

Potato launched last year and today announced a $1 million pre-seed round. The startup also recently shared that it is the first partner in an sintético intelligence program created by wiley, an academic and educational publishing powerhouse.

The company's tuberous name is a nod to the classic experiment where children turn a potato into a battery using household items like pennies, copper wire, and galvanized nails.

“That's really what we're doing,” Edwards said. “We are helping people plan and conduct experiments.”

Edwards, the startup's CEO, has first-hand experience working hard at the research bank. He has a PhD in neuroscience and conducted research at the National Institutes of Health and the University of California, San Diego. He recalled reading magazine articles to develop research protocols.

But there can be inaccuracies in papers and inconsistent approaches, which has created a reproducibility crisis where scientists can't always replicate their own or others' work and get the same results.

Potato uses chat-based generative AI tools built from large language models that are further refined using retrieval augmented generation (RAG). RAG allows them to improve the accuracy of results by incorporating references from scientific literature that will now be expanded to include Wiley publications.

Potato's platform is already being used in laboratories at biotech companies and universities, including the University of Washington; Stanford; Harvard; Massachusetts Institute of Technology, University of California, San Diego; University of California, Berkeley; The Scripps Research Institute; and others.

“The goal is really to enable researchers to quickly distill scientific content and design modified, repeatable experiments so they can accelerate the path to medical advances,” Kosai said.

Kosai is Potato's chief technology officer. He was previously founder and CTO of fintech startup Attunely, which has since closed, and was CTO of startup studio Pioneer Sq. Labs. He also worked for ExtraHop, Marchex and Athleon.

The tool can also critique journal articles, help researchers write articles for publication, and assist in reviewing journal submissions. It can look for deficiencies in protocols, such as weak controls, and help generate ideas about new lines of research.

One hope is that it can find subtle, but potentially revolutionary, tweaks to the protocols. Edwards thought back to his own neuroscience research and realized that swapping one salt for another in a commonly used solution would keep the tissue he was studying alive for twice as long. Information like that can easily go unnoticed in traditional, informal methods of sharing protocols, he said, which include word-of-mouth exchanges and the other way around.

There are other AI research tools available, although many focus on digesting and summarizing published research. Other people working in this space include nonprofit organizations and academic institutions.

One challenge for co-founders is to guard against errors or hallucinations in AI results. They are doing their own fact-checking and users will be directed to the source documents and can refer to the tool for more information.

While the current focus is on biology and medicine, the startup plans to expand its AI assistant to help with research in materials science and chemistry.

Scientists can join the platform for free for limited use and can pay to upload and access additional information and expand their collaborations.

Participants in the seed funding include Axial, Pioneer Sq. Labs, AI2 Incubator and Bow Haber, as well as angel investors such as Cloudera founder Jeff Hammerbacher and ExtraHop founder Jesse Rothstein.



5do">Source link

Leave a Comment