A systematic review and meta-analysis examined artificial intelligence for predicting microsatellite instability from haematoxylin and eosin-stained histopathology in colorectal cancer. Colorectal cancer is the third most commonly diagnosed malignancy and the second leading cause of cancer-related mortality worldwide, accounting for approximately 1.9 million new cases and over 900,000 deaths each year. Microsatellite instability status guides selection of candidates for pembrolizumab- or nivolumab-based therapy, refines prognostic estimates, and identifies patients for Lynch syndrome evaluation. Reference-standard determination relies on polymerase chain reaction-based microsatellite analysis, immunohistochemistry, or next-generation sequencing, but these methods impose substantial demands on laboratory infrastructure. The seminal work of Kather et al. in 2019 demonstrated that deep convolutional neural networks could predict MSI status directly from H&E-stained whole-slide images. The past 24 months have witnessed a paradigm shift including histopathology-specific foundation models, regulatory-grade validation culminating in CE-IVD marking of MSIntuit CRC, and multi-institutional external validation cohorts. Existing systematic reviews predate or only partially cover this paradigm shift.