A platform capable of performing 1,000 chemical experiments per day using artificial intelligence (AI) and robots has been developed, which is expected to accelerate the development of new drugs and materials in the future.

Ulsan National Institute of Science and Technology (UNIST) announced on the 25th that a research team led by Professor Bartosz Grzybowski of the Department of Chemistry has developed an 'AI·Robot-based Automated Experiment Platform' that can automatically perform more than 1,000 chemical experiments per day using artificial intelligence (AI) and robots.

This technology is expected to provide a groundbreaking turning point in the development of new drugs and the discovery of new materials by enabling precise mapping of complex chemical reaction networks and rapid generation of new chemical compounds.

This study was published online in the international academic journal Nature on September 25th, and is generating significant buzz in the fields of chemistry and AI convergence research both domestically and internationally.

Conventional chemical reactions are described by simple equations such as 'A+B=C', but actual reactions take the form of complex networks that produce various results depending on conditions such as temperature, concentration, and time.

The research team simultaneously tested thousands of reaction conditions using a self-developed automated platform and visualized the results in the form of a precise reaction map.

Through this, hidden reaction pathways are discovered and new products that were previously unknown are created.I also confirmed many of them.

A representative example is the "Hantschu pyridine synthesis reaction," first reported 150 years ago, which was precisely analyzed to identify nine new intermediates and products in addition to the seven existing products. This demonstrates that even classical reactions still present uncharted territory.

This platform is also significant in that it can rapidly accumulate experimental data essential for AI learning.

Previously, data acquisition was limited due to manual experiments, but by automatically generating large amounts of reaction data through robots and linking them with AI, the speed of new material discovery can be dramatically increased.

“We plan to accelerate the discovery of new molecules by integrating AI and robotics, and apply this to the development of practical new materials,” said first author Yankai Jia.

Professor Grzybowski emphasized, “The perspective of viewing chemical reactions as networks rather than straight lines will be a major turning point in future chemical research,” and “Automated experiments utilizing AI and robots will significantly increase the efficiency and diversity of chemical synthesis, contributing to the development of new drugs and material innovation in the future.”