Finding Future Solutions with AI

"We are optimistic. We believe we can find new antibiotics and cancer drugs, and new processes for sustainable chemical production," says Jensen.

He is a professor at the Department of Chemistry at the University of Bergen. Together with Professor Bengt Erik Haug and other colleagues at the Department of Chemistry, he focuses on artificial intelligence and robot-based experiments.

The robot facility has attracted significant attention, and recently both the Minister of Research and Higher Education, Oddmund Hoel, and Erna Solberg visited the laboratory to learn more about the facility, the research opportunities it provides, and the plans for the much larger national facility "Norwegian open infrastructure for high-throughput experimentation and scale-up (NorHTE)" which was granted in the Research Council's infrastructure allocation last autumn.

Innovative Methods and Autonomy

Professor Jensen explains: "We actually believe that we can create the next generation of antibiotics and sustainable production processes. Although it is like finding needles in a haystack, we know the needles are there. It's about searching enough and searching smartly."

A key factor in the group's work is the use of artificial intelligence and robotics to perform complex tasks. While it is both useful and relatively easy to get the robot facility to automatically perform a series of similar tasks, the group is working on developing autonomy. In other words, they want the robot itself – based on one or more already performed experiments – to decide which experiments should be done next.

So far, they have managed to get the robot facility to optimize yields in simple processes entirely on its own. They are now working to extend this autonomy to increasingly complex tasks. This will be much easier when the more advanced NorHTE facility is in place sometime in 2026.

In the long term, the group wants to combine virtual, computational searches for new molecules with robot-based synthesis and testing of the most interesting candidate molecules. This would provide a "molecule discovery platform" that can independently predict, synthesize, and test new catalysts and other functional molecules.

Searching for Sustainable Solutions

The group is working on developing new, sustainable ways to produce ammonia, a research that is part of the FME center HyValue. Ammonia does not contain carbon and has great potential as a green fuel, especially in shipping. However, to exploit this potential, the existing production process (called Haber-Bosch), which involves natural gas, must be replaced by a more environmentally friendly process.

To develop this and other processes, the research group uses self-developed methods to design molecules. They have developed the first general method for automatic, so-called "de novo" design of all types of molecules, including catalysts, drugs, and other functional molecules. This method has been experimentally verified and represents a significant milestone in the field.

The group collaborates with international experts and has received significant support from the Research Council of Norway. Their ability to accelerate the discovery of new functional molecules will be further enhanced with the upcoming national infrastructure, NorHTE.

Educating Future Researchers

"There is little knowledge about AI-driven, robot-based experimentation in society or industry. We must encourage Norwegian industry to use these tools. Therefore, we will educate the next generation of researchers to use these fantastic tools for automation and autonomy," says Jensen.

For more information about the research group, you can visit the website or contact Professor Vidar Remi Jensen.