With　the　development　of　the　economy　and　infrastructure　construction,　a　large　number　of　slopes　have　been　formed,　and　the　threat　of　landslides　to　the　safety　of　people＇s　lives　and　property　is　increasing.　Therefore,　it　is　particularly　important　to　monitor　slope　status　effectively.　The　continuous　progress　of　artificial　intelligence　(AI)　has　solved　the　problem　of　slope　safety　and　its　prediction.　However,　there　are　too　many　factors　affecting　the　slope　so　the　predictions　of　the　models　vary.　To　obtain　the　best　prediction　model,　continuous　experiments　are　needed.　Therefore,　we　applied　AI　in　slope　monitoring.　Data　was　collected　and　processed　through　a　system,　and　five　algorithms　were　compared　in　terms　of　performance:　linear　regression　(LR),　deep　learning　(DL),　decision　tree　(DT),　support　vector　regression　(SVR),　and　multi-layer　perceptron　(MP).　Slope　settlement　was　predicted　based　on　pore　water　pressure,　rainfall,　cracks,　and　humidity.　The　root　mean　square　error　and　experimental　results　were　analyzed　based　on　each　model,　and　the　deep　learning　algorithm　model　was　incorporated　into　the　prediction　system.　The　research　results　indicated　that　slope　hazards　could　be　predicted　with　the　lowest　root　mean　square　error　to　prevent　hazards.　©　2023　IEEE.