U.S. Department of Homeland Security Public Sector Generative AI Deployment Handbook ()

Key Chapter Title List

1. Message from the Secretary and Chief AI Officer
2. Introduction: The DHS Generative AI Pilot Projects
3. Pilot One: Enhancing Search and Summarization Using Large Language Models to Strengthen Investigative Leads
4. Pilot Two: Assisting Local Governments in Developing Hazard Mitigation Plans
5. Pilot Three: Creating New Training Opportunities for Immigration Officers
6. Enhancing Mission-Oriented Generative AI Use Cases
7. Coalition Building and Effective Governance
8. Tools and Infrastructure
9. Responsible Use and Trustworthiness Considerations
10. Measurement and Monitoring
11. Training and Talent Acquisition
12. Usability Testing and Other Feedback Mechanisms

Document Introduction

The rapid development and commercialization of Generative Artificial Intelligence (GenAI) present significant opportunities and challenges for the U.S. Department of Homeland Security (DHS) and public sector organizations at all levels. In response to the requirements of Executive Order 14110 for the responsible deployment of AI technology, DHS has taken the lead in launching Generative AI pilot projects to explore its potential for enhancing mission capabilities while steadfastly upholding core principles of privacy protection, civil rights, and civil liberties.

This handbook systematically synthesizes the practical experience from DHS's three major Generative AI pilots, constructing a structured framework for the responsible adoption of Generative AI technology within the public sector. The three pilots focus respectively on: enhancing the search and summarization functions of investigative reports using Large Language Models to strengthen Homeland Security investigative leads; leveraging AI to assist local governments in efficiently developing hazard mitigation plans, thereby improving community resilience; and providing dynamic, personalized training for immigration officers to optimize the interview process for refugee and asylum applications. These pilots are all positioned as augmenting employee work rather than replacing human labor, emphasizing replicability and cross-departmental applicability.

The handbook provides practical guidance across seven core dimensions: clarifying that Generative AI deployment must be precisely aligned with the organizational mission, designing narrowly scoped, high-value pilot projects; emphasizing the critical role of senior leadership support and cross-functional coalition building, establishing a robust AI governance structure; guiding organizations to assess existing tools and infrastructure, making informed choices between commercial, open-source, or open-weight models; integrating responsible use and trustworthiness throughout, identifying and mitigating potential risks related to privacy, security, and bias; establishing scientific indicator systems and monitoring mechanisms to continuously track pilot effectiveness; enhancing employee AI literacy through targeted training and attracting specialized technical talent; and building iterative feedback mechanisms to integrate user and stakeholder input for product optimization.

As an authoritative guide for Generative AI deployment in the public sector, this handbook not only summarizes DHS's key experiences in AI governance, technology integration, and risk management but also provides practical tools such as pilot proposal templates and sample indicator dashboards. The handbook emphasizes that the successful integration of Generative AI requires multi-stakeholder collaboration among government, industry, academia, and civil society. The ultimate goal is to fully unlock the transformative potential of Generative AI through responsible innovation, enhancing public sector service capabilities to better serve the American people.