EcosystemSUMMIT & RESEARCH
Evidence library / Advanced Solutions for Development Delivery
AI for development · SELECTED MOMENT · 0:47

Unsuccessful projects can reveal mistakes and conditions hidden by success-only evidence.

Likely speaker
Sophie PascaleAI Growth Specialist · World Bank, Development Impact AI LabLinkedIn ↗
Unidentified speaker

Free short preview · The problem is fragmentation

Speaker turns in this passage

Select a time to jump to that part of the passage. Speaker changes are approximate.

Sophie PascaleAI Growth Specialist · World Bank, Development Impact AI LabLinkedIn ↗
Unidentified speaker
Sophie PascaleAI Growth Specialist · World Bank, Development Impact AI LabLinkedIn ↗

Unsuccessful projects can reveal mistakes and conditions hidden by success-only evidence.

View citation

Attributed contributors: Sophie Pascale. “Unsuccessful projects can reveal mistakes and conditions hidden by success-only evidence..” Advanced Solutions for Development Delivery. ECOSYSTEM Summit, Barcelona, 16 September 2026. Session time 15:48–16:35. https://cs-ecosystem.commonshare.workers.dev/evidence/E-ed742a678703

Interpretation and context.

Analyst synthesis. The recording establishes that a claim was made. It does not independently establish the claimed effect.

practitioner argument

Unsuccessful projects can reveal mistakes and conditions hidden by success-only evidence.

Contributors
Sophie PascaleAI Growth Specialist · World Bank, Development Impact AI LabLinkedIn ↗
Unidentified speaker

Limit: Failures must actually be reported to enter the corpus.

negative case learningevaluation

Where it connects.

Credible judgement requires methods and context appropriate to the question, with failure and missing voices visible.

The corpus contains studies, practitioner accounts, demonstrations, proposals and questions. These establish different things. Local knowledge and negative cases can reveal what a generic measure misses.

Follow the evidence →