Jaya Kusuma Radja Modjo Pengarang DEWI I N BILI BORA Dosen Pembimbing 2 IDA NYOMAN TRI DARMA PUTRA Dosen Pembimbing 1 Fransiskus Bustan Ketua Penguji Dewi I N Bili Bora Penguji 2 Ida Nyoman Tri Darma Putra Penguji 1

This study aims to analyze translation errors produced by Google Translate. The method used is descriptive qualitative by collecting data in the form of text in the source language (Bahasa Indonesia) and the translation results in the target language (Bahasa Inggris). Data are analyzed based on the categories of translation errors according to the theory of Molina and Albir (2002), including grammatical, lexical, semantic, and pragmatic errors. The results of the study show that Google Translate still has limitations in translating idiomatic phrases, cultural contexts, and complex sentence structures. The most dominant errors occur at the semantic and pragmatic levels, where the machine is unable to capture the nuances of meaning that depend on context. This study concludes that although Google Translate can help with basic translation, its accuracy is still limited so human editing is needed for more natural and precise results. Keywords: Error analysis, Google Translate, machine translation, semantics, pragmatics.