PDF Intake (PDFI) for myUSCIS

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

2025-07-23 (day precision)Source: 2025-07-23T00:00:00

Withheld from public reporting

Not reported

Mission

Topic area

Government Benefits Processing

Operational functions

Document processing and extraction Derived by keyword rules; see methodology

Problem the AI is intended to solve

Scanned PDFs submitted through MyUSCIS must be validated against form-specific business rules related to both the overall document and the contents of specific fields. Constructed a service that can process a scanned input document and return all information pertinent to these validation rules in a consistent structure (JSON) to a user-facing ELIS microservice. The GenAI powered library utilizes Amazon Bedrock – Anthropic Claude 3.7 Sonnet V1 Foundation Model to extract data from PDF forms. The service provides ability to submit forms online through MyUSCIS UI to Lockbox instead of via mail.

Expected benefits

Develop a service that can extract relevant fields from a scanned PDF submitted through MyUSCIS and build a JSON as an output to the ELIS microservice. The new service will utilize AWS Bedrock provided foundation model. It is an engineering solution that minimizes development time to add new forms or form revisions with high accuracy.

System outputs

The output of this AI system is structured information about the validation rules applied to the input form as well as the extracted contents of filled fields on the form, presented in a JSON format readable by both humans and machines that is consistent with existing ELIS databases.

Technology

AI classification

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

System name(s)

USCIS Electronic Immigration System (ELIS)

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)

Analytica and DV United

Vendors (standardized)

None on the standard list

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

Data and privacy

Involves PII

Yes

Privacy Impact Assessment

www.dhs.gov ↗

Authorization to Operate

Yes

Demographic features

Not reported

Training and evaluation data

During development, this system is evaluated using both manually created synthetic data (i.e. filled PDF forms with annotated contents) and production data (scans of forms submitted previously through Lockbox as scanned TIF files). The underlying pretrained foundation model supplied by Bedrock service is used as-is with no further training or fine-tuning.

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