Prof. Martin Huber, from the University of Fribourg, and Dr. David Imhof, from the Swiss Competition Commission, have developed an artificial intelligence-based method to detect cartels and price fixing among firms bidding for contracts in public tenders (e.g. road construction contracts). The researchers utilized so-called deep learning algorithms to analyze the joint bidding behavior of firms and to learn to distinguish between cartels and competitive patterns. The method was tested using procurement data from Switzerland and Japan, where it identified roughly 95% of firms (i.e., 19 out of 20 firms) correctly as either cartel participants or competitors. This innovative approach has the potential to help competition agencies detect and combat cartels and other collusive activities. The method has been published in the International Journal of Industrial Organization and is available open access at: https://doi.org/10.1016/j.ijindorg.2023.102946.