JobFetch

AI Trainer - Data Annotation Specialist (Safety)

Pay
Pay not listed
Where
Remote
Type
Contract
Eligibility
Country eligibility not stated. Check with the provider before applying.
Listing
Last verified Nov 2, 2026

Apply opens the provider’s application page. You apply there yourself: we never submit applications, and opening the page doesn’t mean you’ve applied. Nobody can promise you’ll be hired.

Some Apply links are referral links. If you apply through them, the provider may pay us a commission. Referral disclosure

Explore weekly job alerts for your skills

Overview

Evaluate AI-generated responses against defined safety policies as a data annotator.

A non-coding role for candidates with strong communication skills, native-level English proficiency, excellent attention to detail, close-reading ability, sound judgment, and the ability to write clear, concise evaluations.

Responsibilities

  • Evaluate AI-generated responses against defined safety policies and sub-policies.
  • Review single-turn and multi-turn conversations for safety, accuracy, and policy compliance.
  • Assign consistent 1-5 ratings based on established evaluation criteria.
  • Write clear, concise, and evidence-based justifications for each rating.
  • Identify subtle safety concerns involving self-harm, violence, abuse, and other sensitive topics.

Requirements

  • Strong interest in data annotation and AI evaluation with excellent English proficiency.
  • Excellent reading comprehension and close-reading skills.
  • Ability to write clear, concise, and objective evaluations.
  • Excellent attention to detail and awareness of subtle nuances.
  • Strong judgment and ability to follow detailed safety policies.
  • Ability to work accurately under time constraints.
  • Comfort with reviewing sensitive content involving self-harm, violence, and abuse.
  • Familiarity with US-specific references, systems, and terminology.

Preferred

  • Experience in data annotation, content moderation, or policy review.
  • Background in trust and safety, AI training, or response evaluation.
  • Experience in editorial review, proofreading, compliance, or quality assurance.
  • Familiarity with structured rating systems and evaluation rubrics.
  • Experience reviewing single-turn and multi-turn conversations.

Engagement and pay

  • Pay per task; no amount stated. Average handling time is approximately 75 minutes per task.
  • Application requires an updated resume and LinkedIn profile link; selection follows internal and client review.

Eligibility

  • Country or residence limits are not stated.

Skills

  • analysis
  • content moderation
  • trust & safety