Overview
Trust & Safety Evaluator conduct adversarial testing and safety evaluation of generative AI features. The role involves crafting queries, evaluating the safety of generated content, and providing critical feedback to improve safety systems.
Responsibilities
- Design and test challenging queries and attack scenarios to expose potential misuse.
- Simulate realistic conversations and evaluate the safety of generated content.
- Assess biases, cultural and linguistic nuances, and subtle safety risks.
- Provide actionable feedback to guide safety evaluations.
- Maintain high accuracy while completing tasks within time constraints.
- Create creative content to demonstrate and communicate findings.
Qualifications
- Fluency in English (South Africa) and deep cultural awareness.
- Understanding of cultural references and industry context.
- Familiarity with youth behavior, slang ...