AI Terms of Service
Addendum to Salary Master Services Agreement
Artificial Intelligence Addendum (“AI Terms”)
1. Applicability and Relationship to the MSA
1.1 Applicability. These AI Terms govern Customer’s access to, use of, and reliance on AI Features. These AI Terms apply only to AI Features identified in an applicable Order Form or otherwise enabled by Salary for Customer. If Customer does not purchase, enable, or use AI Features, these AI Terms do not apply. If applicable, these AI Terms are incorporated by reference to the Master Services Agreement (“MSA”) between Salary and Customer. Except as expressly provided in these AI Terms, nothing in these AI Terms limits, modifies, or replaces the parties’ rights or obligations under the MSA. Any defined terms not specifically defined in the AI Terms retain the meaning that they have in the MSA.
1.2 Purpose of these AI Terms. AI Features are designed to assist Customer by generating analyses, recommendations, summaries, insights, or other content based on Customer’s Prompts, Customer Data, Salary Data, and other information made available through the Services. Because AI Features present risks and considerations that differ from traditional software functionality, these AI Terms allocate the parties’ respective rights, responsibilities, and risks associated with the use of AI Features.
2. Defined Terms
This section defines terms used throughout the AI Terms.
2.1 “AI Features” means the artificial intelligence-enabled functionality made available by Salary through the Services, including generative, predictive, analytical, conversational, recommendation, summarization, classification, workflow automation, or other AI-powered capabilities that are identified in an applicable Order Form, enabled by Salary, or otherwise made available to Customer.
2.2 “Prompt” means an instruction, request, query, question, text, data, file, image, content, or other information submitted by Customer or an Authorized User to an AI Feature for processing.
2.3 “AI Output” means a response, recommendation, summary, analysis, classification, prediction, insight, ranking, visualization, text, image, content, or other material generated or returned by an AI Feature in response to a Prompt.
2.4 “AI Providers” means Salary’s third-party providers of large language models, foundation models, machine learning models, AI infrastructure, inference services, or other artificial intelligence technologies used by Salary in providing AI Features.
3. Customer Responsibilities
3.1 Customer Control and Responsibility. Customer determines whether, when, and how AI Features are enabled and used within its organization. Customer is solely responsible for configuring, supervising, and controlling its use of AI Features and for determining whether AI Features are appropriate for Customer’s intended use cases. Customer acknowledges that AI Features are decision-support tools designed to assist Customer and are not intended to replace Customer’s business judgment, professional judgment, human review, internal controls, or legal compliance obligations.
3.2 Responsibility for Prompts. Customer is solely responsible for all Prompts submitted to AI Features by Customer or its Authorized Users. Customer represents and warrants that it has all rights, permissions, authorizations, and consents necessary to submit each Prompt and to authorize Salary and its AI Providers to process the Prompt as contemplated by the MSA, these AI Terms, and the applicable OF. Customer further represents and warrants that its Prompts do not violate applicable law, infringe or misappropriate the rights of any third party, or breach any contractual, fiduciary, confidentiality, or other legal obligation owed to another person or entity.
3.3 Prohibited Data. Customer shall not knowingly submit Prohibited Data to AI Features except where expressly authorized in the applicable Documentation, OF, or DPA. If Customer becomes aware that Prohibited Data has been submitted to an AI Feature without authorization, Customer shall promptly notify Salary and reasonably cooperate with Salary in identifying, securing, and removing such information.
3.4 Review and Validation of AI Output. Customer acknowledges that AI Output is generated using probabilistic technologies and may be incomplete, inaccurate, inconsistent, outdated, biased, or otherwise unsuitable for Customer's intended purpose. Before relying upon or acting on any AI Output, Customer shall independently review, validate, and evaluate the AI Output using appropriate human judgment and any additional review appropriate under the circumstances. Customer remains solely responsible for determining whether AI Output is accurate, appropriate, and suitable for its intended use.
3.5 Human Oversight. Customer is solely responsible for determining whether applicable law, regulation, professional standards, contractual obligations, Customer’s internal policies, or the nature of Customer’s intended use requires human review or oversight before AI Output may be relied upon or implemented. Where such review or oversight is required, Customer shall ensure that appropriately qualified personnel review the AI Output before any resulting decision or action is taken. Salary shall have no responsibility or liability arising from Customer’s failure to implement or maintain human review or oversight.
3.6 Customer Decisions. Customer is solely responsible for all decisions, actions, and omissions that are made using, informed by, or based upon AI Output. Without limiting the foregoing, Customer is solely responsible for all employment, compensation, organizational, hiring, promotion, pay equity, compliance, legal, financial, operational, and other business decisions made using AI Features or AI Output. Customer acknowledges that AI Features provide information and recommendations only and do not make decisions on Customer’s behalf.
3.7 Compliance with Applicable Law. Customer is solely responsible for determining whether and how AI Features may be used in compliance with applicable law. Customer is solely responsible for determining whether its use of AI Features requires disclosures, notices, human oversight, testing, documentation, auditing, recordkeeping, risk assessments, or other compliance measures under applicable law or Customer's own internal governance requirements. Salary does not provide legal advice regarding Customer’s use of AI Features and does not determine whether Customer’s intended or actual use complies with applicable law.
4. Salary’s AI Commitments
4.1 Purpose of AI Features. Salary makes AI Features available to assist Customer by generating analyses, recommendations, summaries, insights, classifications, predictions, and other AI-generated content. AI Features are intended to supplement Customer’s use of the Services and are provided subject to the limitations and disclaimers in these AI Terms.
4.2 Use of Customer Data. Except as expressly permitted under the MSA, the DPA, or applicable OF/SOW, Salary will not use Customer Data, Prompts, or AI Output to train or fine-tune shared foundation models or large language models for the benefit of other customers. Nothing in this Section limits Salary’s rights under the MSA to use Usage Data, Anonymous Data, Feedback, or other information that Salary is expressly permitted to use under the MSA.
4.3 AI Providers. Salary may utilize one or more AI Providers in connection with the operation of AI Features. Salary may replace, add, or discontinue AI Providers from time to time as part of its ordinary operation and evolution of the Services, provided that doing so does not materially diminish the AI Features purchased by Customer during the applicable Term, except where reasonably necessary to address security, legal, operational, or technical requirements.
5. AI Output
5.1 Nature of AI Output. AI Output is intended to assist Customer by providing analyses, recommendations, summaries, insights, classifications, predictions, or other informational content. Because AI Features use probabilistic technologies, AI Output may vary in response to similar Prompts and may contain inaccuracies, omissions, inconsistencies, outdated information, or other errors. Customer acknowledges that AI Output is inherently probabilistic and should be evaluated in the context of Customer's intended use.
5.2 AI Output Disclaimer. AI Output is provided “as is” and “as available.” Salary does not represent or warrant that AI Output:
is accurate, complete, current, or error-free;
is suitable for Customer's intended purpose;
complies with applicable law, regulatory requirements, or industry standards;
is free from inaccuracies, hallucinations, bias, omissions, or other defects;
will produce consistent, repeatable, or predictable results; or
will satisfy Customer's business, legal, operational, compliance, or regulatory objectives.
AI Output may reflect assumptions, patterns, or statistical relationships derived from training data or Customer inputs and should not be interpreted as a statement of fact unless independently verified by Customer.
5.3 No Professional Advice. AI Output is provided for informational and assistive purposes only and does not constitute legal, accounting, tax, employment, compensation, financial, regulatory, or other professional advice. Nothing generated through an AI Feature should be construed as a substitute for independent professional advice.
5.4 Ownership of AI Output. As between the parties, and subject to Salary’s ownership of the Services, AI Features, Salary Data, software, models, algorithms, methodologies, Documentation, and all other underlying intellectual property, Customer owns the AI Output generated specifically in response to Customer's Prompts. Nothing in these AI Terms transfers to Customer any ownership interest in Salary’s artificial intelligence models, software, algorithms, system prompts, training methodologies, reference data, Salary Data, or other intellectual property used to develop, operate, or provide AI Features. Customer’s rights to use AI Output are subject to the restrictions contained in the MSA and these AI Terms.
6. AI-Specific Use Restrictions
6.1 AI-Specific Permitted Use. Customer may use AI Output only for Customer’s internal business purposes unless the applicable OF expressly provides otherwise.
6.2 AI-Specific Prohibited Uses. In addition to the restrictions contained in the MSA, Customer shall not, and shall not permit any Authorized User or person acting under Customer's authority to:
use AI Features to generate content or take actions in violation of applicable law;
attempt to manipulate, circumvent, disable, or defeat the safeguards, safety mechanisms, usage restrictions, or operational controls incorporated into AI Features;
engage in prompt injection, jailbreak attempts, adversarial testing, model extraction, or other techniques intended to discover, manipulate, bypass, or alter the operation of AI Features beyond their intended functionality;
use AI Features to identify, discover, extract, infer, reconstruct, or obtain Salary's system prompts, model configurations, proprietary methodologies, algorithms, weights, training techniques, or other non-public aspects of the AI Features;
use AI Features, AI Output, Salary Data, or other information obtained through AI Features to develop, train, benchmark, fine-tune, improve, validate, or evaluate any artificial intelligence model, machine learning model, large language model, dataset, or other product or service that competes with Salary or substitutes for the Services;
use automated tools, scripts, bots, or similar technologies to submit high-volume Prompts or otherwise interact with AI Features in a manner intended to circumvent usage limits or normal operation;
represent AI Output as having been independently generated by Customer or by a human where such representation would be false or misleading; or
use AI Features in any manner that materially interferes with the security, availability, integrity, or performance of the AI Features or the Services.
6.3 No Circumvention of Human Oversight. Customer shall not use AI Features for any purpose that intentionally avoids, circumvents, or replaces human review or oversight where such review or oversight is required by applicable law, regulation, contractual obligation, Customer's internal policies, or the intended use of the AI Output. Nothing in this Section imposes an independent obligation on Salary to determine whether such human review or oversight is required.
6.4 Salary’s Right to Protect AI Features. If Salary reasonably believes that Customer or an Authorized User has violated these AI Terms or that Customer’s use of AI Features presents a material security, legal, operational, or reputational risk to Salary, the Services, another customer, or a third party, Salary may investigate the suspected conduct and may suspend, limit, disable, or terminate access to the affected AI Features to the extent reasonably necessary to protect the Services or enforce these AI Terms. Where commercially reasonable, Salary will provide Customer with notice of the applicable issue and an opportunity to cure before suspending AI Features. Salary may act immediately without prior notice where reasonably necessary to protect the Services, respond to an actual or suspected security incident, comply with applicable law, or prevent material harm. Suspension under this Section is in addition to, and not in lieu of, rights or remedies available to Salary under the MSA.
7. AI Data Processing
7.1 Application of the MSA and DPA. Except as expressly provided in these AI Terms, Customer Data processed through AI Features is governed by the MSA and, where applicable, the DPA. Nothing in these AI Terms modifies the parties’ respective rights and obligations regarding Customer Data, Personal Data, Usage Data, Anonymous Data, or Confidential Information under the MSA or the DPA.
7.2 Processing of Prompts and AI Output. Salary may process Prompts, AI Output, and related information as reasonably necessary to:
provide and operate AI Features;
authenticate users and process Customer requests;
maintain, support, secure, monitor, troubleshoot, and improve the performance and reliability of the Services;
investigate suspected misuse, fraud, or security incidents;
comply with applicable law; and
exercise Salary’s rights and perform its obligations under the MSA and these AI Terms.
7.3 Operational Logs and Telemetry. Salary may collect, retain, and use operational, diagnostic, security, audit, and usage logs relating to AI Features in accordance with its ordinary business practices and applicable law. Such logs may include information reasonably necessary to operate, secure, troubleshoot, investigate, audit, and improve AI Features, provided that Salary’s use of Customer Data remains subject to the MSA and the DPA.
7.4 Model Training. Except to the extent Customer expressly agrees otherwise in writing, Salary will not use Customer Data, Prompts, or AI Output to train or fine-tune shared foundation models or large language models for the benefit of other customers. Nothing in this Section limits Salary’s rights under the MSA with respect to Usage Data, Anonymous Data, Feedback, or other information that Salary is expressly permitted to use under the MSA.
8. Indemnification
8.1 Customer Indemnification of Salary. Without limiting Customer's indemnification obligations under the MSA, Customer shall defend, indemnify, and hold harmless Salary, its Affiliates, and their respective officers, directors, employees, agents, successors, and assigns from and against any third-party claim arising out of or relating to Customer’s:
Prompts or other information submitted to AI Features;
use or misuse of AI Features or AI Output;
violation of these AI Terms;
violation of applicable law in connection with AI Features;
failure to obtain any authorization, consent, notice, disclosure, approval, or human oversight required for Customer’s use of AI Features or AI Output;
infringement, misappropriation, or violation of a third party’s intellectual property, privacy, publicity, confidentiality, contractual, or other rights arising from Customer’s use of AI Features or AI Output; or
submission of Prohibited Data.
Customer’s obligations under this Section supplement, and do not replace, Customer’s indemnification obligations under the MSA.
8.2 Breach of the AI Terms. Any breach of these AI Terms constitutes a material breach of the MSA. Salary may exercise any rights or remedies available under the MSA arising from such breach, including suspension or termination of AI Features or the affected Services, in addition to any other rights or remedies available at law or in equity.
9. Risk Allocation
9.1 AI-Specific Allocation of Risk. Without limiting any disclaimer, limitation of liability, or allocation of responsibility contained in the MSA, Salary shall not be liable for claims, losses, damages, liabilities, penalties, fines, regulatory actions, or other consequences arising from or relating to:
Customer’s use of AI Features or AI Output in violation of applicable law;
Customer’s failure to satisfy any legal, regulatory, contractual, or internal governance requirement applicable to Customer’s use of AI Features, including any requirement relating to human oversight, disclosure, documentation, testing, auditing, monitoring, or approval;
Customer’s reliance upon AI Output; or
Customer’s implementation of recommendations, analyses, classifications, predictions, summaries, or other AI Output.
9.2 Application of the MSA. The disclaimers, limitations of liability, exclusions of damages, liability cap, and all other risk allocation provisions contained in the MSA apply to these AI Terms and to Customer’s access to and use of AI Features. Nothing in these AI Terms expands Salary’s warranties or liability beyond those expressly set forth in the MSA.
10. Updates to these AI Terms
Salary may modify these AI Terms from time to time to reflect changes to AI Features, applicable law, security practices, technology, or operational requirements. If a modification materially reduces Customer’s rights or materially increases Customer’s obligations with respect to AI Features during the then-current Term, Salary will provide Customer with reasonable prior notice. Unless otherwise required by applicable law or expressly agreed in writing, updated AI Terms will become effective upon the effective date specified in the notice. Customer’s continued use of AI Features after the effective date of the updated AI Terms constitutes acceptance of the updated AI Terms.