The State of AI in Government: An Analysis of Adoption Patterns at the USDA

EasyChair Preprint 16007

Abstract

The promise of generative AI (GenAI) has captured the global imagination, yet its journey into the heart of government remains a mystery. While the private sector sprints toward an AI-driven future, the public sector seems to be taking a different, more deliberate path. This study pulls back the curtain on this digital divide by examining 89 AI projects from the USDA's public inventory. We found that the fanfare surrounding GenAI masks a quiet reality: conventional technologies like Natural Language Processing (NLP), Computer Vision (CV), and Traditional Machine Learning (ML) overwhelmingly dominate the landscape, with GenAI representing a mere fraction of total cases. This study contributes to IS research by highlighting the cautious trajectory of AI adoption in government and discussing implications for the diffusion of emerging technologies, accountability, and digital innovation ecosystems.

Keyphrases: Artificial Intelligence (AI), Digital Government, Digital Transformation, Generative AI (GenAI), Information Systems (IS), technology diffusion

@booklet{EasyChair:16007,
  author    = {Hrishitva Patel},
  title     = {The State of AI in Government: An Analysis of Adoption Patterns at the USDA},
  howpublished = {EasyChair Preprint 16007},
  year      = {EasyChair, 2026}}