The paper summarizes the first international steganalysis challenge called BOSS (Break Our Steganographic System). The challenge aimed to assess the robustness of steganographic systems and encourage research in steganalysis. A new content-adaptive steganographic algorithm called HUGO was developed for the challenge. HUGO was designed to be highly undetectable by minimizing the distortion between cover and stego images. The challenge involved two image databases: BOSSBase and BOSSRank. BOSSBase was used for developing steganalyzers, while BOSSRank was used for evaluation. The challenge faced a cover-source mismatch, where the training and testing images came from different sources, forcing participants to design steganalyzers robust to this discrepancy. The challenge also highlighted the importance of reducing false positive rates and improving the accuracy of steganalyzers. The results showed that three teams achieved high accuracy, with Hugobreakers scoring the highest. The challenge also demonstrated the importance of clustering analysis and the potential benefits of combining multiple steganalyzers. The paper concludes that BOSS has stimulated research in steganalysis and has identified important challenges for future work, including improving the robustness of steganalyzers and reducing false positive rates.The paper summarizes the first international steganalysis challenge called BOSS (Break Our Steganographic System). The challenge aimed to assess the robustness of steganographic systems and encourage research in steganalysis. A new content-adaptive steganographic algorithm called HUGO was developed for the challenge. HUGO was designed to be highly undetectable by minimizing the distortion between cover and stego images. The challenge involved two image databases: BOSSBase and BOSSRank. BOSSBase was used for developing steganalyzers, while BOSSRank was used for evaluation. The challenge faced a cover-source mismatch, where the training and testing images came from different sources, forcing participants to design steganalyzers robust to this discrepancy. The challenge also highlighted the importance of reducing false positive rates and improving the accuracy of steganalyzers. The results showed that three teams achieved high accuracy, with Hugobreakers scoring the highest. The challenge also demonstrated the importance of clustering analysis and the potential benefits of combining multiple steganalyzers. The paper concludes that BOSS has stimulated research in steganalysis and has identified important challenges for future work, including improving the robustness of steganalyzers and reducing false positive rates.