Insights
Quick Learnings on AI Safety
Category
Lessons Learned - Partnering to Pilot Responsible AI in Colombia
Stress-Testing AI at the Humanitarian Frontier: Findings from Signpost Colombia's First AI Simulation Round
Info Pa'lante Colombia: Framing a Case Study in Scaling Responsible AI
How Evidence Shaped aprendIA
InfoDigna: How IRC Mexico Is Using AI to Expand Reach Without Compromising Safety
aprendIA Northeast Nigeria Pilot: Early Performance Data Shows Promise
Designing Safe Risk Escalation in Client-facing Agentic AI Systems
Testing Before Trusting: The Red-Teaming Process Behind aprendIA's Safe Deployment
aprendIA: AI-Powered Support for Teachers in Resource-Constrained Settings
Dungeons & Dragons Character Sheets for AI Evaluators: Building AI Evaluators That Know Why a Response is Good
Beyond Awareness: How Personalized Information Empowers Refugees to Recognize Exploitation and Seek Help
Conserving Token Usage and Reducing AI Inference Costs in Production
A Practical Approach to Humanitarian AI Safeguarding and Guardrails
Behind the Scenes: Orchestrating AI for Humanitarian Response
Executive Summary: Signpost AI Information Assistant Pilot Project
Training AI for Trust: Insights from Cuéntanos El Salvador's Chatbot Journey
![[BtB | PO Diary] Reflections from A Protection Officer](https://images.squarespace-cdn.com/content/v1/65a80959361431491cb9dc19/1719335910852-DQQIZMDE00MNJQCDCVMM/Collages+%286%29.jpg)
[BtB | PO Diary] Reflections from A Protection Officer
New Year, New Horizons: Signpost AI’s Vision for 2025
Transparency and Accountability in AI: A Panel Discussion
