Taxpayer Services Chatbot

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

The output is not presumed to be high-impact and is not used as the principal basis for significant decisions/actions

Start date

2023-03-14 (day precision)Source: 3/14/2023

Withheld from public reporting

NoSource: a) No

Mission

Topic area

Service Delivery

Operational functions

Chatbot or virtual assistant Derived by keyword rules; see methodology

Problem the AI is intended to solve

This project implemented the Taxpayer Services chatbot, which provides self-service Frequently Asked Questions (FAQ) to taxpayers and international taxpayers regarding refunds, status of refunds, amended returns, Employer Identification Number (EIN), Earned Income Tax Credit (EITC), change of address, identity theft, extension to file topics. The Intent Engine, using Natural Language Processing (NLP) and AI algorithms classifies taxpayers' utterances (questions) into intents (requests). The answers provided to the taxpayers are not created by Generative AI. The responses of the chatbot and business logic is predetermined by content owners. This AI tool aims to classify/navigate to a correct predetermined response.

Expected benefits

This chatbot provides means for the taxpayers to receive information without need to speak with an Agent. The benefits include reduced taxpayer burden as well as increased internal IRS operational efficiencies.

System outputs

The NLP algorithm functions by classifying user inputs into predefined topics using advanced natural language processing techniques. It evaluates the input against a set of trained models and assigns a response based on a confidence score calculated by the algorithm. Only responses associated with these pre-established topics are provided, ensuring accuracy and consistency. The algorithm does not generate new or ad-hoc answers outside the defined topics, making it reliable for controlled environments where precision and adherence to predetermined content are critical.

Technology

AI classification

Natural language processingSource: Natural Language Processing: AI that processes, interprets, and shares information in human language.

System name(s)

Not reported

Custom-developed code

No

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)

Not reported

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

No

Privacy Impact Assessment

Not reported

Authorization to Operate

Yes

Demographic features

k) None of the above

Training and evaluation data

Learning set of utterances-to-intent maps.

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