Gender bias in machine translation: an analysis of Google Translate in English and Spanish

Abstract

Although Google has attempted on several occasions to remove gender bias from its free online translation service, it still tends to exhibit predominantly masculine options and shows a tendency towards perpetuating or exaggerating sexist stereotypes, which adds to other flaws like a failure to notice text formality, typos and nuances. We analyse the gender bias of Google Translate between English and Spanish by entering a number of gender-invariable Spanish nouns whose referent’s gender was unknown due to the omission of pronouns and other particles. The results show a strong gender bias which needs to be removed by training machine translation software to infer semantic gender from pronouns and word terminations, and applying gender-tagging in multilingual corpora.