New platform aims to enhance AI research accuracy
AI is making its way into more and more areas of life and work. In some areas though, particularly scientific research, it's vitally important to ensure the accuracy of results.
Norwegian company Iris.ai has developed a method to measure the factual accuracy of AI-generated content -- testing precision and recall, fact tracing, and extraction.
It's launching Researcher Workspace 1.0, a new platform that uses natural language processing (NLP) to accelerate scientific research, putting factual accuracy first. Researchers can access a comprehensive AI tool suite powered by Iris.ai's technology, including a brand new chat feature, to search, explore, and extract information from swathes of scientific documents all in one place without compromising privacy.
In a survey of 500 corporate R&D workers, Iris.ai found that only 22 percent trust popular AI tools such as ChatGPT. Early tests on the Researcher Workspace chat functionality show that features soon to be introduced have reduced AI hallucinations down to single-figure percentages. In order to maximise privacy, the models only interact with user content for the specific session or operation requested, never learning or storing document information.
Using knowledge graphs and a universal metric named WISDM, designed by the company, Iris.ai tests context similarity to confirm factual accuracy.
Anita Schjøll Abildgaard, CEO and co-founder of Iris.ai, says:
The Researcher Workspace 1.0 is the product of seven years of research and development, based on the latest in AI, machine learning, and NLP technology. This is a game-changer for scientific researchers, and we’re grateful to our clients and customers who’ve been invaluable in our journey along the way.
The Researcher Workspace 1.0’s ability to help researchers navigate, evaluate, and draw insights from thousands of research documents in hours, not months, can accelerate the pace of discovery and innovation and help us tackle some of the biggest problems of our age: our clients are working on topics like climate change, plastic pollution, human health, and the spread of disease.
You can find out more on the Iris.ai site.
Photo credit: Peshkova / Shutterstock