Artificial Intelligence based Arabic-to-English machine versus human translation of poetry: An analytical study of outcomes

This study discusses the discrepancies between a machine translation and human of Arabic poems to English. The dataset comprises two poems, Nothing Note, One Day. Each these have English translations achieved by one or more translators working in unison ensure close original. researcher then generated Artificial Intelligence (AI)-based Machine Translation (MT) versions conducted critical linguistic examination outputs. Results indicated that MT fails capture cultural context for instance, poem, protagonist is uprooted his salary changed "pounds," use different adjectives creates meaning than translation's conjunctive pairs same lyrical continuance adjectival antonyms, which not MT. poem's also lost MT, e.g., "the dust-dunned street" "dusty street," "like time would walk with him" "on ground, as if he was walking, but wasn't passing." Finally, establishes protagonists, while does not. In second analysis Day, limitations are stark capturing socio-cultural poetry. comparing points out latter failed nuances including figurative language, historical references, genre poem. conclusion, unable apply beyond its database lacks ability understand poetry created can therefore, be good tool