Please use this identifier to cite or link to this item:
https://dair.nps.edu/handle/123456789/5391
Title: | Leveraging Generative AI for Validating the Quality of DoD Acquisition Packages and Contract Documents |
Authors: | Samantha Nangia Tom Wardwell Randall Mora Carlos Parada Jr. |
Keywords: | Defense acquisition Generative AI NIPR GPT LLM FAR compliance DFARS compliance PPMAP NMCARS |
Issue Date: | 2-May-2025 |
Publisher: | Acquisition Research Program |
Citation: | APA |
Series/Report no.: | Acquisition Management;SYM-AM-25-316 ;SYM-AM-25-403 |
Abstract: | The Department of Defense (DoD) contracting process requires rigorous validation to ensure regulatory compliance, accuracy, and completeness. This paper explores the integration of NIPR GPT, a secure generative artificial intelligence (AI) model, to enhance the efficiency and reliability of the Acquisition and Contracting package validation. Deployed in a DoD-approved environment, NIPR GPT is a Government R&D Platform for GenAI models and applications serving as a comprehensive AI research and development platform featuring retrieval augmented generation. NIPRGPT enables model evaluation, shared workspaces, and secure document processing workflows, in our use case, we used it to automate key tasks such as compliance checks against FAR/DFARS/NMCARS/Local Policy Language, clause verification, contract risk identification, and data consistency validation. The proposed framework enables contracting officers to upload documents, select validation tasks, and receive detailed, actionable reports. NIPR GPT is able to leverage fine-tuned training on DoD-specific datasets to identify missing clauses, resolve ambiguities, and flag high-risk elements. By automating labor-intensive tasks, the system is able to reduce human error, accelerate processing, and ensure compliance with regulatory and policy requirements. The model is implemented within an IL-4 environment to address security concerns, with robust encryption protocols and access controls to safeguard sensitive data. Audit logging provides transparency, ensuring outputs can be reviewed and verified. A case study using a significant Aircraft procurement demonstrates the practical application of this framework. NIPR GPT identified missing compliance language and clauses, flagged ambiguous deliverable descriptions, and recommended corrective actions, streamlining the package approval process. This integration of AI into DoD workflows illustrates its potential to modernize procurement practices, improve accuracy, and maintain compliance in a highly regulated environment. This abstract highlights the transformative role of generative AI in supporting DoD contracting officers by providing reliable, secure, and efficient tools for package validation. |
Description: | SYM Paper / SYM Presentation |
URI: | https://dair.nps.edu/handle/123456789/5391 |
Appears in Collections: | Annual Acquisition Research Symposium Proceedings & Presentations |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
SYM-AM-25-316.pdf | SYM Paper | 810.1 kB | Adobe PDF | View/Open |
SYM-AM-25-403.pdf | SYM Presentation | 5.73 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.