AI Safety Content Evaluator (Arabic/English Required)

About OpenTrain

OpenTrain is a centralized job board for data-labeling and AI-training roles. We aggregate live contract openings from many AI teams and labeling platforms so you can discover relevant opportunities in one place.

Creating an OpenTrain account is free, and applying takes only a few minutes — we help match skilled people to short- and long-term projects in AI safety, content moderation, transcription, annotation, and more.

About AI Training and Safety Work

AI training (also called data labeling, annotation, or human feedback work) is the human side of building AI systems. For LLMs, that often means evaluating model outputs, producing example responses, and guiding models away from unsafe or incorrect behavior.

Safety-focused roles involve reviewing potentially explicit or disturbing content and documenting model failures so engineers can fix risks and improve system behavior.

The Role

We are hiring an AI Safety Content Evaluator who is fluent in Arabic and highly proficient in English to review and rate AI-generated text with an emphasis on safety.

This fully remote, hourly contractor position centers on evaluating reasoning quality, annotating safety concerns, and producing clear, policy-aligned feedback so models avoid toxic, unsafe, or adversarial outputs.

What You'll Do

• Review and rate AI-generated responses in English and Arabic, focusing on safety, factuality, and clarity.
• Annotate outputs for safety categories such as hate, harassment, sexual content, suicide/self-harm, violence, bias, illegal activity, malicious code, and misinformation.
• Identify and document adversarial prompts or red-team style attacks that produce unsafe model behavior.
• Provide written justifications and clear explanations for ratings, especially in ambiguous or borderline cases.
• Follow strict safety guidelines and documentation standards while maintaining consistent policy application.
• Work with text-only data and evaluation interfaces labeled as OTHER (platform/tool specifics provided during onboarding).

Requirements

• Near-native or native Arabic proficiency in reading and writing.
• Minimum C1 English proficiency in reading and writing.
• Bachelor’s degree or higher in Communications, Linguistics, Psychology, Law/Policy, Security Studies, or equivalent professional experience.
• Proven experience in Trust & Safety, content moderation, policy enforcement, risk operations, investigations, or safety evaluation.
• Required hands-on LLM red teaming experience, including identifying adversarial prompts and documenting unsafe model behaviors.
• Strong knowledge of safety domains: hate/harassment, sexual content, suicide/self-harm, violence, bias, illegal goods/services, malicious activities/malicious code, and misinformation.
• Ability to apply written safety policies consistently and explain decisions clearly in ambiguous cases.
• Comfortable reviewing explicit, toxic, violent, sexual, or psychologically disturbing content as part of daily work.
• Practical experience using tools such as Perplexity, Gemini, ChatGPT, or similar AI systems.
• Prior experience with AI data training, annotation, or evaluation workflows is preferred.

Who Should Apply

This position is a good fit for mid-level safety specialists, experienced content moderators, policy analysts, or researchers who are fluent in Arabic and strong in English and who enjoy methodical, impactful evaluation work.

Ideal candidates are careful writers, comfortable with sensitive material, and experienced at documenting edge cases and adversarial attacks against language models.

Compensation and Logistics

Employment type: Contractor, Part-time. Time requirement: 20+ hours per week.

Pay: Hourly, typically $25/hr with an advertised range of $15–$40 USD per hour. Exact rate depends on project and experience.

Work is fully remote and worldwide; you will work with text data and evaluation/rating tasks (label types: EVALUATION_RATING, TEXT_GENERATION).

How It Works / How To Apply

Apply through OpenTrain by creating a free account and submitting your profile and any requested samples or documentation.

If selected, you'll receive project-specific onboarding, access to the labeling/evaluation interface, and safety guidelines. Expect to document decisions carefully and follow prescribed annotation workflows.