The 8th AI City Challenge focused on applying artificial intelligence in areas like retail, warehouses, and Intelligent Traffic Systems (ITS), offering significant research opportunities. The 2024 edition featured five tracks, attracting 726 teams from 47 countries. Track 1 involved multi-target multi-camera (MTMC) people tracking, with enhanced camera counts and 3D annotations. Track 2 focused on traffic safety analysis through detailed video captions. Track 3 required classifying driver actions in naturalistic driving scenarios. Track 4 explored fish-eye camera analytics using the FishEye8K dataset. Track 5 aimed to detect motorcycle helmet rule violations. The challenge used two leaderboards to showcase methods, with participants setting new benchmarks. The datasets included MTMC people tracking, Woven Traffic Safety (WTS), SynDD2, FishEye8K, and Bike Helmet Violation Detection. Evaluation metrics varied by track, with Track 1 using HOTA scores, Track 2 using BLEU-4, METEOR, ROUGE-L, and CIDEr, Track 3 using average activity overlap, Track 4 using F1 score, and Track 5 using mean Average Precision (mAP). Teams employed advanced models and techniques, including YOLO-based models, transformers, and ensemble methods. Notable results included high HOTA scores in Track 1, detailed captions in Track 2, accurate activity recognition in Track 3, and effective object detection in Track 4. Track 5 saw a top mAP of 0.4860. The challenge highlighted the importance of practical, scalable applications in AI, with a focus on improving safety and efficiency in retail and ITS. Future research directions include addressing class imbalance, enhancing online tracking methods, and exploring spatial-temporal relationships in visual language models. The 8th AI City Challenge marked a milestone with new tasks and enhanced datasets, showcasing the potential of AI in real-world applications.The 8th AI City Challenge focused on applying artificial intelligence in areas like retail, warehouses, and Intelligent Traffic Systems (ITS), offering significant research opportunities. The 2024 edition featured five tracks, attracting 726 teams from 47 countries. Track 1 involved multi-target multi-camera (MTMC) people tracking, with enhanced camera counts and 3D annotations. Track 2 focused on traffic safety analysis through detailed video captions. Track 3 required classifying driver actions in naturalistic driving scenarios. Track 4 explored fish-eye camera analytics using the FishEye8K dataset. Track 5 aimed to detect motorcycle helmet rule violations. The challenge used two leaderboards to showcase methods, with participants setting new benchmarks. The datasets included MTMC people tracking, Woven Traffic Safety (WTS), SynDD2, FishEye8K, and Bike Helmet Violation Detection. Evaluation metrics varied by track, with Track 1 using HOTA scores, Track 2 using BLEU-4, METEOR, ROUGE-L, and CIDEr, Track 3 using average activity overlap, Track 4 using F1 score, and Track 5 using mean Average Precision (mAP). Teams employed advanced models and techniques, including YOLO-based models, transformers, and ensemble methods. Notable results included high HOTA scores in Track 1, detailed captions in Track 2, accurate activity recognition in Track 3, and effective object detection in Track 4. Track 5 saw a top mAP of 0.4860. The challenge highlighted the importance of practical, scalable applications in AI, with a focus on improving safety and efficiency in retail and ITS. Future research directions include addressing class imbalance, enhancing online tracking methods, and exploring spatial-temporal relationships in visual language models. The 8th AI City Challenge marked a milestone with new tasks and enhanced datasets, showcasing the potential of AI in real-world applications.