GenAI for Document Summarization and Content Generation

Overview

Development stage

DeployedSource: c) Deployed – The use case is being actively authorized or utilized to support the functions or mission of an agency.

High-impact designation

Not high-impactSource: c) Not high-impact

Impact justification

Not reported

Start date

2025-03-01 (day precision)Source: 2025-03-01T00:00:00

Withheld from public reporting

Not reported

Mission

Topic area

Administrative Functions

Operational functions

Summarization Derived by keyword rules; see methodology

Problem the AI is intended to solve

Workforce enablement and efficiency through the implementation of Generative AI / Large Language Models (LLM)

Expected benefits

Improving Operational Efficiency: Automating information retrieval, reducing manual review time, and streamlining workflows will lead to significant time savings and increased productivity for CBP personnel. This translates to cost savings and allows agents to focus on higher-priority tasks. Enhancing Decision-Making: Providing quick and accurate access to relevant information will improve the quality and consistency of decisions across the agency.

System outputs

Generative LLM applications deployed in a stand alone capacity or embedded in existing systems that will allow users to upload, search, summarize documents, conduct advanced searches, and identify similar language in documents and receive natural language-style outputs in response to their prompts.

Technology

AI classification

Generative AISource: Generative AI: AI that generates new or synthetic content (e.g., images, videos, audio, text, code).

System name(s)

Not reported

Custom-developed code

Yes

Public source code

Not reported

Sourcing

How it was built

Contract and in-houseSource: c) Developed with both contracting and in-house resources

Vendor (as reported)

Meta, OpenAI, Google, Anthropic

Vendors (standardized)

OpenAIGoogleAnthropicMMeta

A vendor is the supplier named by the agency. It does not identify the underlying model or AI technology.

Data and privacy

Involves PII

No

Privacy Impact Assessment

Not reported

Authorization to Operate

No

Demographic features

Not reported

Training and evaluation data

The commercial LLMs used for this use case were trained using a diverse range of publicly available data, including text from books, articles, websites, and other sources and data types.

Federal Data Catalog

Not reported

Governance

0 of 8 minimum-practice questions answered. A blank answer means the agency reported nothing; it does not mean the practice is absent.

Potential impacts and how they were identified

Not reported