Brendan Raftery, Yasmine Abuzeid, Selina Tedesco, Silvia Garuti, Andrea Noble, and Björg Flygenring Finnbogadottir

In current systems of aid, relief is distributed based on hypotheses and outdated knowledge. While most organizations collect data about disasters, they rarely gather insights from people experiencing disasters directly. Earshot, my communication tool, democratizes information for both those affected by natural disasters and for the organizations and governments that serve them. At the center of my solution is an AI-driven chatbot named Quinn. Quinn can answer questions from people affected by a natural disaster because of Earshot’s robust database that is built by text messages and phone calls from previous survivors. Question such as, How do I apply for a grant to repair my home? Which pharmacies nearby have the prescription I need? are tackled by Earshot. Earshot uses machine learning to convert this database into actionable insights for governments and other organizations to deliver better, faster, and more accurate aid. The business model of Earshot will be a subscription-based nonprofit one, where our customers (aid organizations/governments) receive actionable insights related to disaster relief efforts. The structure as a nonprofit would free the company from the pressures of hyper growth while qualifying it for philanthropic investment and relevant grants. End users would include anyone impacted by a disaster and looking for information or having questions.

Project Link