From Evidence to Decisions in the Age of AI: Strengthening Evaluation Use and Trust
Webinar (em inglês) | Online
-
Organizado por:
Learn Adapt Manage (LAM)
Sobre o evento
This interactive session explores how evaluation can maintain trust and credibility in an AI-driven information environment. As AI transforms how evidence is generated, synthesized, and communicated, the challenge shifts from producing more evidence to designing evaluations around the decisions they are intended to inform and ensuring findings are translated and used responsibly. Participants will engage in discussions and reflections on practical ways to strengthen the use of evidence, human judgment, and decision-making systems in their contexts.
Orador/a
| Nome | Título | Biography |
|---|---|---|
| Florence Randari | Monitoring, Evaluation, and Learning Advisor & Founder, Learn Adapt Manage (LAM) | Florence Randari is a Monitoring, Evaluation, and Learning Advisor and founder of Learn Adapt Manage (LAM). She supports development programs to strengthen evidence use, program learning, and adaptive management, with a focus on turning data into better decisions. |
| Mary Mwikali | Evidence to Action Advisor at IRC | Mary Mwikali is an Evidence to Action Advisor at IRC, where she works at the intersection of evidence and learning. She works on evidence synthesis, translation, and the use of AI to strengthen how evidence is produced and applied. She writes the ABCs of Evidence a newsletter on AI and evidence use. |
Resumo
The session concluded that maintaining trust and credibility in evaluation in the age of AI requires more than faster evidence generation and synthesis. Participants reflected on the importance of designing evaluations around the decisions they are intended to inform, clarifying evidence needs early, and ensuring appropriate levels of rigor for different contexts. Discussions emphasized that while AI can support evidence synthesis, translation, and communication, it cannot replace human judgment, contextual understanding, and collective interpretation. The webinar reinforced that trustworthy evaluation depends on intentional decision systems, responsible AI use, and meaningful human involvement throughout the evidence-to-decision process.
Participants were encouraged to continue reflecting on how evaluations are designed and used within their organizations, particularly in relation to decision-making processes and responsible AI use. Follow-up actions include strengthening intentional design for evidence use, improving collective interpretation and ownership of findings, and developing clearer approaches for integrating AI into evidence synthesis and translation workflows. Participants were also invited to continue the conversation through professional networks, share practical experiences from their contexts, and engage with additional learning resources shared after the session.