AI Machine Translation Jeopardizing Afghan Asylum Claims

 

These tools are on the rise, increasingly used by for-profit government contractors and aid organizations that work with refugees and migrants.

Not only does AI translation cost translators their jobs, but it is quite literally endangering asylum cases. Machine translations of Pashto and Dari, in particular, are riddled with errors that have introduced confusion into already complex immigration processes, and led to the rejected asylum claim of at least one Afghan refugee in the past year.

At Respond we frequently see the U.S. government pointing to minor technicalities on asylum applications to deny them; mistranslated details are literally detrimental to asylum claims. As our Respond translator, Muhammed Yaseen says: “If we rely on machine translation, at least 40% of our decision-making on asylum applications for refugees would be incorrect”.

The skilled trauma-informed human interpreters of the Respond Afghan Languages have already translated 900 cases. Since January 2023, we’ve supported 188 Afghans in winning their asylum claims. Talented human translators are essential and irreplaceable. 

Please help us amplify this critical article!

SHARE: Instagram / Twitter / Article


 
Previous
Previous

Respond Crisis Translation on PBS NewsHour: Machine translation is endangering asylum claims

Next
Next

和伤害中文使用者的语言暴力作斗争 Combatting Language Violence Against Chinese Speakers