AI Safety Content Evaluator (Arabic/English Required)

Project Overview / Instructions

Applicants must have a bachelor’s degree or higher in a relevant field, such as linguistics, psychology, law, or security studies, or equivalent experience. Native or near-native proficiency in Arabic is required, with at least C1-level English for high-precision safety labeling and documentation. Experience in Trust & Safety, content moderation, policy enforcement, risk operations, investigations, or safety evaluation work is required. LLM red teaming experience and strong knowledge of nuanced safety domains—including hate, harassment, explicit content, and misinformation—are necessary. Project contributors will curate and annotate safety-focused training examples, review and score model responses, identify policy gaps, and document reasoning failures in English and Arabic. The role involves continuously stress-testing AI models to ensure compliance with safety policies, proposing decision rule improvements, and maintaining consistent annotation standards across multiple reviewers. Content may include explicit or psychologically distressing materials.