Request for Proposal Independent Red Teaming and AI Safety Evaluation

• Conduct independent red teaming and AI safety evaluation of Girl Effect's WAZZII chatbot and generative AI systems serving adolescent girls across Kenya
• Lead rigorous assessment of retrieval-grounded responses, safety guardrails and failure modes to identify risks and vulnerabilities in the girl-facing experience
• Analyse compliance with Girl Effect's Ethical AI Guidelines and evaluate safeguarding protocols, including the Failure Mode Register and escalation procedures
• Deliver a comprehensive evaluation report with findings, recommendations and evidence-based insights on AI safety performance over a 10-week engagement
• Maintain strict independence throughout the assignment and confirm no involvement in building, operating or commercially benefiting from the systems evaluated

─────────────────────────────────────

This summary is automatically generated for quick reference. For the complete and authoritative job description, please view the original posting.