AI-Enabled Workflow Automation
Description
Office of Immigration Litigation (OIL) faces significant backlogs, with repetitive or frivolous claims slowing progress and foreign-language evidence creating bottlenecks. This initiative uses Natural Language Processing for triage and automate translation to accelerate case processing. It aligns with M-25-21’s innovation and governance principles by improving efficiency while preserving human oversight.
Detailed example
Classifications, Automated Translations, Recommendations: case priority classifications, language translations, processing recommendations.
AI / analytics pattern
Natural Language Processing: AI that processes, interprets, and shares information in human language.
Automation level / stage
a) Pre-deployment – The use case is in a development or acquisition status.
Expected benefit
(1) Reduces immigration case backlogs, ensures consistent treatment of claims, and improves responsiveness to federal courts. (2) Builds on OIL case management systems, United States Citizenship and Immigration Services data feeds, and DOJ translation contract costs.(3) Reduced backlog size; fewer attorney hours per case; translation accuracy benchmarks met. Increase in fraudulent/frivolous and repetitive claims.
Controls / human review
ATO: Not reported; PIA: Not published