AI Agents – Terms of Service

These Unit21 AI Agent Online Terms of Service (“Terms”) specifically govern the use and provision of the "Unit21 AI Agents" feature (the “AI Agent”) offered by Unit21, Inc. (“Vendor”). These Terms are incorporated by reference into, and supplement, the Master Services Agreement (“MSA”), Statement of Work (“SOW”), and any other agreements between Vendor and Customer (as defined in the applicable SOW or MSA). 

‍

In the event of any conflict or inconsistency between these Terms and any other agreement, including but not limited to the MSA or SOW, these Terms shall prevail solely with respect to the use, operation, and provision of the AI Agent. Capitalized terms not defined herein have the meanings set forth in the Agreement.

‍

By accessing or using the AI Agent, Customers agree to be bound by these Terms, as they may be updated from time to time. The Vendor reserves the right to modify these Terms at its discretion, with such changes being effective upon notice to Customers or upon publication on the Vendor’s platform.

‍

1. Purpose and Scope. The AI Agent is an optional feature designed to support Customer's anti-money laundering (AML) compliance and fraud mitigation workflows through orchestration of actions and responses powered by third-party large language models (LLMs).

‍

2. Third-Party Model Disclaimer. Customer acknowledges that the AI Agent leverages third-party LLMs and infrastructure. Vendor does not control or develop these foundational models and makes no representations or warranties regarding their accuracy, reliability, behavior, or performance.

‍

3. Use Restrictions. Customer agrees not to use the AI Agent in any manner that: (a) Violates applicable laws, regulations, or guidance (including AML, data protection, export control, or financial regulatory laws); (b) Inputs, uploads, or transmits information that is classified, subject to banking secrecy or export restrictions, or otherwise requires special handling without all necessary authorizations; (c) Attempts to reverse engineer, probe, or extract proprietary elements of the underlying models; (d) Uses the AI Agent to train, develop, or benchmark competing machine learning models without Vendor's prior written consent. Customer acknowledges that outputs from the AI Agent are not intended to serve as the sole basis for regulatory filings or compliance decisions without appropriate human validation and assumes all risk associated with such use.

‍

4. Customer Responsibilities. Customer is solely responsible for: (a) Ensuring that inputs submitted to the AI Agent are accurate, lawful, and appropriate; (b) Validating, interpreting, and reviewing AI Agent outputs before taking any action in reliance on them; (c) Configuring any automated or downstream decision-making triggered by AI Agent outputs; (d) Complying with all applicable regulatory obligations arising from its use of the AI Agent.

‍

5. Output Limitations. Customer acknowledges that: (a) Outputs from the AI Agent are generated by probabilistic machine learning models and may contain errors, omissions, or inaccuracies; (b) Vendor disclaims all liability arising from Customer's reliance on AI Agent outputs; and (c) the AI Agent does not constitute legal, compliance, or regulatory advice.

‍

6. Regulatory Disclaimer. Vendor does not represent or warrant that the AI Agent complies with or satisfies any regulatory requirement. The AI Agent is provided solely as a tool to support Customer’s internal processes. Customer must independently determine the suitability of the AI Agent for its regulatory obligations.

‍

7. Model Updates and Evolution. Vendor may modify, upgrade, or replace the underlying models, orchestration logic, or user interface of the AI Agent at any time. Customer is responsible for testing and validating continued use following any such updates.

‍

8. Security and Data Use. Vendor will not use Customer Data for model training. Customer acknowledges that inputs and outputs may be transmitted to third-party model providers solely for the purpose of generating responses. 

‍

9. Subprocessors. Customer acknowledges and consents to Unit21’s use of third-party service providers (subprocessors) to support the operation of the AI Agents. Without limiting the foregoing, Customer expressly consents to the engagement of the following subprocessors:

• Amazon Web Services, Inc., including the use of Amazon Bedrock services for access to foundation models under AWS management;
• Google Cloud Platform, including Google Search services (search and enhancement services).

‍

10. Service Availability. The AI Agent is provided on an "as-is" and "as-available" basis. No uptime or service-level guarantees apply to the AI Agent, and it may be suspended or discontinued at Vendor’s discretion.

‍

11. Suspension and Termination. Vendor may suspend or terminate Customer’s access to the AI Agent upon notice if: (a) Customer breaches these Terms; (b) use of the AI Agent poses a regulatory, legal, or security risk to Vendor as determined by Vendor in its sole discretion; or (c) Vendor discontinues the AI Agent in its discretion.

‍

12. Governing Terms. These Terms are subject to, and incorporated into, the Agreement between Unit21 and Customer. In the event of any conflict between these Terms and the Agreement, these Terms shall govern solely with respect to Customer’s use of the AI Agents.