AI Trainer, Safety Data Annotation Specialist (English)

You will review AI-generated text and conversations against safety policies and related rules. The work includes sensitive content involving possible self-harm, violence, and abuse, so careful and objective judgment is important.

You will rate responses using established criteria, explain your decisions with concise evidence-based reasoning, and identify subtle safety concerns in complex conversations.

• Evaluate AI-generated responses for safety, accuracy, and policy compliance.
• Review both single-turn and multi-turn conversations.
• Assign ratings from 1 to 5 using defined criteria.
• Write clear, objective reasons for your ratings.
• Identify subtle safety concerns in complex content.
• Review sensitive material involving possible self-harm, violence, and abuse.

What It Pays and Takes

This is an entry-level, part-time contractor role. The listing does not provide a pay rate.

• Workload: 20 or more hours per week.
• Location: Open worldwide.
• Language: Professional fluency in English.
• Strong written communication and reading comprehension.
• Close-reading ability, attention to detail, and sound judgment.
• Ability to follow detailed safety policies accurately under time constraints.
• Comfort reviewing sensitive content.
• Familiarity with US-specific references, systems, and terminology.
• Helpful experience includes data annotation, content moderation, policy review, trust and safety, AI training, response evaluation, editorial review, proofreading, compliance, or quality assurance.
• Familiarity with structured rating systems, evaluation rubrics, and conversation review is helpful.

How It Works

Apply on OpenTrain with your resume, then complete the application on the hiring site.

About AI Training Work

AI training is the human work behind systems that learn from examples, including reviewing and rating model responses. Careful reviewers help make these systems safer, more accurate, and more useful.