AI Safety and Policy Evaluator

Role overview Help evaluate how AI handles sensitive image-edit requests. You’ll use policy guidance and careful judgement to identify safety issues and explain your decisions clearly. What you’ll be doing • Classify prompts and model responses against a defined safety policy. • Explain your decisions in clear, self-contained written rationales. • Identify borderline cases, failures and gaps in the evaluation guidelines. • Where the project brief calls for it, design test prompts or compare alternative model outputs. Who this could suit People with experience in content moderation, trust and safety, policy analysis, AI evaluation or related analytical work. The work involves written material about sensitive topics, including self-harm, sexual content, violence and harassment. What you’ll need A degree or equivalent experience in a relevant field, strong analytical judgement, precise writing, and the ability to apply nuanced policy categories consistently. Experience with adversarial testing or prompt design is needed; previous content moderation or AI-safety work is strongly preferred. You must provide your own computer and reliable internet connection. You don’t have to match every requirement exactly. We welcome applicants with different backgrounds, levels of experience and transferable skills. Pay and working arrangements Remote, project-based independent contractor work. Fees are agreed for each statement of work according to its scope and required expertise. The current project term is one to two weeks, and contractors set their own schedules. Required training is separately compensated under the project terms. What happens next Apply through Find Jobs in AI with your profile and CV. We’ll review your application and, if you’re a good fit, we’ll be in touch about the next steps.