MSPai Servicing Digital Chatbot and IVA

Artificial Intelligence Training Data Transparency Disclosure

Version: Pilot | March 2026

The purpose of this disclosure is to provide a high-level summary of the datasets used in the training of this generative artificial intelligence system.

DATASET 1:

Item	Respone
Developer Name	Black Knight Servicing Technologies, LLC
Sources or owners of the datasets	Internally created test data
How does the dataset further the intended purpose of the artificial intelligence system or service?	The test dataset is representative of information that would be in a production environment, such that AI system performance and accuracy can be evaluated.
Number of data points included in the dataset(s)	Approximately 300
Description of the types of data points within the datasets	For dataset(s) with label, types of labels used: Types of data labels include those related to: Funding sources (i.e., bank account name, funding source account name, last 4 digits of bank account number routing transit numbers, etc.) Autopay data fields (i.e., loan number, draft frequency type, draft frequency amount, next draft amount, draft delay days, etc.) Loan information data fields (i.e., Draft indicator, ARM indicator, escrow account balance, unpaid principle balance, loan type, loan maturity date, etc.) Escrow data fields (i.e., escrow account balance, anticipated escrow expenses, insurance policy type description, tax disbursement amount, prior year net taxes paid, etc.) Loan support data fields (i.e., HELOC lien position, unpaid principle balance, annual percentage rate, scheduled total payments, cash out amount, home equity eligibility indicator, etc.) Payment and payment eligibility data fields (i.e., one-time draft fee amount, current bill due date, one-time late fee amount, next bill date. Delinquent payment count, fee code, fee description, minimum payment amounts due, debit card service fee indicator, etc.) Paperless statement data fields (i.e., statement type and delivery preference) For dataset(s) without labels, general characteristics of the dataset(s): N/A
Does the dataset(s) include data protected by copyright, trademark, patent?	No
Is the dataset(s) entirely within the public domain?	No
Was the dataset(s) purchased or licensed?	N/A
Does the dataset(s) include personally identifiable information¹ ?	No
Does the dataset(s) include aggregate consumer information² ?	No
Was there any cleaning, processing, or other modification to the dataset(s)? If so, please explain the intended purpose of these efforts.	No
Time period during which the data in the datasets were collected. Please indicate whether the data collection is ongoing.	01/2017; data collection is not ongoing.
Dates the dataset(s) were first used during the development of the artificial intelligence system	08/2025
Does the artificial intelligence system use, or has it used, synthetic data generation in its development? If yes, please provide a description of the functional need or desired purpose for such use.	No

DATASET 2:

Item	Response
Developer Name	Black Knight Servicing Technologies, LLC
Sources or owners of the datasets	Internally created test data
How does the dataset further the intended purpose of the artificial intelligence system or service?	This dataset provides the source of knowledge for responding to general knowledge mortgage questions.
Number of data points included in the dataset(s)	Approximately 110
Description of the types of data points within the datasets	For dataset(s) with label, types of labels used: Generic mortgage-related questions and associated generic answers. For dataset(s) without labels, general characteristics of the dataset(s): N/A
Does the dataset(s) include data protected by copyright, trademark, patent?	No
Is the dataset(s) entirely within the public domain?	No
Was the dataset(s) purchased or licensed?	N/A
Does the dataset(s) include personally identifiable information¹ ?	No
Does the dataset(s) include aggregate consumer information² ?	No
Was there any cleaning, processing, or other modification to the dataset(s)? If so, please explain the intended purpose of these efforts.	No
Time period during which the data in the datasets were collected. Please indicate whether the data collection is ongoing.	General mortgage questions and answers (“Knowledge Base”) were created in November 2025. Knowledge Base questions and answers are updated and added to on an as- needed basis.
Dates the dataset(s) were first used during the development of the artificial intelligence system	Testing of the initial list of Knowledge Base questions and answers started in November 2025.
Does the artificial intelligence system use, or has it used, synthetic data generation in its development? If yes, please provide a description of the functional need or desired purpose for such use.	No

1 “PII” means information that identifies, relates to, describes, is reasonably capable of being associated with, or could reasonably be linked, directly or indirectly, with a particular consumer or household. See Cal. Civ. Code §1798.140(v)

2 “Aggregate Consumer Information” means information that relates to a group or category of consumers, from which individual consumer identities have been removed, that is not linked or reasonably linkable to any consumer or household, including via a device. “Aggregate consumer information” does not mean one or more individual consumer records that have been deidentified. See Cal. Civ. Code §1798.140(b)