Arabic AI Safety Evaluation Expert

About OpenTrain

OpenTrain is the #1 platform for finding and building careers in AI training and data labeling. OpenTrain AI recruits contributors for specialized projects, helping you discover meaningful work, build a credible AI training profile, and grow a long-term freelance career.

• Remote contractor opportunity
• Worldwide availability
• Part-time schedule with a default commitment of 7 hours per week
• Hourly compensation of $28 to $32

About AI Safety Evaluation Work

AI safety evaluation is part of the human work behind modern AI systems. Contributors review model behavior, test difficult interactions, and provide structured judgments that help improve how AI handles sensitive or potentially harmful content.

• Work directly with prompts and written conversations
• Use human judgment to identify unsafe or adversarial behavior
• Help shape how advanced AI systems respond to sensitive topics
• Apply language and cultural expertise to improve evaluation quality

The Role

OpenTrain AI is seeking an Arabic AI Safety Evaluation Expert to help improve the safety of advanced AI models. You will use native or near-native Arabic fluency, cultural understanding, and careful written reasoning to evaluate sensitive topics and identify problematic or adversarial interactions.

Prior AI or machine learning experience is not required. Training is provided for the workflow, making this an entry-level opportunity for candidates who meet the language, education, and judgment requirements.

• Native or near-native Arabic fluency
• Business-level written English
• Remote contractor position
• Open globally, with Saudi Arabia and the wider MENA region preferred but not required

What You'll Do

You will create expert-level Arabic prompts across sensitive subject areas and apply structured guidelines to classify prompts and conversations. Your evaluations will require close attention to wording, context, escalation, and the potential risks of dual-use information.

You will document the reasoning behind your judgments and flag patterns that could indicate adversarial intent or unsafe model behavior.

• Write expert-level prompts in Arabic
• Evaluate sensitive topics and model interactions
• Classify prompts and conversations using structured guidelines
• Identify adversarial phrasing and escalation patterns
• Review sensitive and dual-use information carefully
• Document the reasoning behind evaluation decisions

Requirements

This role requires strong Arabic writing and evaluation ability, business-level written English, and sound judgment when reviewing sensitive content. A bachelor's degree completed or in progress is required.

You should be able to follow detailed guidelines consistently, explain written judgments clearly, and maintain a high level of attention to detail.

• Native or near-native Arabic fluency for writing and evaluating sensitive prompts
• Business-level written English
• Bachelor's degree completed or in progress
• Strong written reasoning
• Close attention to detail
• Ability to classify prompts and conversations using structured guidelines
• Sound judgment around sensitive and dual-use information
• Ability to identify adversarial phrasing and escalation patterns

Helpful Background

Experience reviewing, grading, or red-teaming written or technical content can support success in this role. Background in trust and safety, content moderation, policy evaluation, or adversarial testing is also useful, though prior AI or machine learning experience is not required.

• Written or technical content review
• Content grading or red-teaming
• Trust and safety
• Content moderation
• Policy evaluation
• Adversarial testing
• Arabic cultural context and careful written-content evaluation

Schedule and Compensation

This is a remote, part-time contractor opportunity with a default commitment of 7 hours per week. The compensation rate is $28 to $32 per hour, and the role is open to applicants worldwide.

• Contractor and part-time employment type
• Default commitment: 7 hours per week
• Pay: $28 to $32 per hour
• Worldwide opportunity
• Saudi Arabia and wider MENA applicants preferred but not required

Build Your AI Training Career

AI training and data labeling are fast-growing ways to work in technology from anywhere. OpenTrain gives contributors one place to manage opportunities, show relevant experience, and build a durable portfolio in a field that helps shape how AI systems behave.

• Create a free OpenTrain account
• Build a profile around your Arabic language and evaluation expertise
• Apply to AI training opportunities in minutes
• Develop experience in safety evaluation and human feedback work