Towards Automated Scoping of AI for Social Good Projects
This addresses a critical resource-intensive task for AI practitioners and domain experts in social good initiatives, though it appears incremental as it applies existing LLM capabilities to a specific bottleneck.
The paper tackles the laborious problem scoping bottleneck in AI for Social Good projects by proposing a Problem Scoping Agent that uses an LLM to generate project proposals, demonstrating through blind review and AI evaluations that it produces proposals comparable to expert-written ones.
Artificial Intelligence for Social Good (AI4SG) is an emerging effort that aims to address complex societal challenges with the powerful capabilities of AI systems. These challenges range from local issues with transit networks to global wildlife preservation. However, regardless of scale, a critical bottleneck for many AI4SG initiatives is the laborious process of problem scoping -- a complex and resource-intensive task -- due to a scarcity of professionals with both technical and domain expertise. Given the remarkable applications of large language models (LLM), we propose a Problem Scoping Agent (PSA) that uses an LLM to generate comprehensive project proposals grounded in scientific literature and real-world knowledge. We demonstrate that our PSA framework generates proposals comparable to those written by experts through a blind review and AI evaluations. Finally, we document the challenges of real-world problem scoping and note several areas for future work.