Prosper Global's VERA cuts Sudan research from six days to two

Prosper Global, formerly Mercy Corps, works in about 30 countries with communities facing conflict, climate change and economic hardship, and its crisis analysts' research could previously take days to weeks. VERA is the organization's attempt to compress that front end of the work: a multi-agent engine that fans out across ReliefWeb, the conflict tracker ACLED and local news, then has a final agent assemble the findings.

The early results Eubank cites are modest in scope but concrete. Beyond the Sudan agricultural report, an employee with no security background produced an election security report in Colombia in a few hours, and VERA, launched in June, also runs in the Democratic Republic of Congo and Ukraine. Eubank stressed that the risks and contexts differ in each place, which is why he favors a measured and slow rollout rather than a fast one.

The design choice matters as much as the speed. Eubank frames VERA as a way to tailor the research process and to add further resources later, giving a curated workflow that earlier iterations of generative AI did not offer — while deliberately leaving the judgment with researchers. Whether that balance holds as the tool spreads to more contexts is likely the open question, though Eubank's own emphasis on caution suggests the organization intends to move deliberately.