AI agents cut critical research time for humanitarian teams in conflict zones

Humanitarian groups in conflict zones are discovering that agentic AI can significantly reduce the time spent on research between a crisis and an informed response. Cloudera Inc. has designed its AI strategy with flexibility in mind, and a nonprofit partner is now testing this approach on the ground. However, the goal is to provide analysts with a faster starting point for their work, not to completely automate research, according to Aaron Eubank, director of data science at Prosper Global, formerly Mercy Corps.

Eubank emphasized that VERA, an AI system developed in collaboration with Cloudera, is designed to keep researchers in control. "It's not the final report that's going to be pushed out to the humanitarian community and teams, but rather a starting point that analysts can use to continue to iterate on their own research," Eubank explained during an exclusive broadcast on theCUBE, SiliconANGLE Media's livestreaming studio.

Multi-agent AI is speeding up humanitarian research, with VERA achieving early results in conflict zones. Prosper Global, which operates in around 30 countries with communities facing conflict, climate change, and economic hardship, built VERA using Cloudera's multi-agent research engine, which operates on Amazon Web Services and leverages models from Anthropic PBC.

The agents gather information from sources such as ReliefWeb, conflict tracker ACLED, and local news, with a final agent assembling the findings. For instance, in Sudan, VERA reduced desk research for an agricultural report from five to six days to just two, saving approximately $1,500. In Colombia, an employee without security expertise produced an election security report in a matter of hours.

VERA is currently operational in Sudan, Colombia, the Democratic Republic of Congo, and Ukraine. Eubank stressed the importance of adopting a measured and cautious approach as they implement this technology in different regions.