AI Terms

1. General

These Stonly AI Terms and Conditions (the "AI Terms") supplement the applicable Stonly Terms and Conditions (the "Terms of Service") between the Customer and Stonly, which are incorporated by reference. These AI Terms apply specifically to features of the Stonly platform that are powered by artificial intelligence or machine learning models ("Stonly AI").
In case of any conflict between the Terms of Service and these AI Terms as it relates specifically to Stonly AI, the AI Terms shall govern. By using Stonly AI, individually and/or as an authorized representative of the legal entity on whose behalf you use Stonly AI ("You" or "Customer"), you confirm that you have read and understood these AI Terms and agree to be bound by them. If you do not agree with all of these AI Terms, you may not use Stonly AI.
A violation of the AI Terms constitutes a violation of the Terms of Service. Capitalized terms not defined herein have the meaning given to them in the Terms of Service.

2. Scope of Stonly AI

Stonly AI encompasses the following categories of features, as made available to the Customer through the Platform:

1. AI-Powered Chatbot. An automated conversational agent that provides responses to end-user questions based on the Customer's knowledge base and configured data sources.
2. AI-Assisted Content Creation. Tools that assist Users in drafting, editing, summarizing, or improving knowledge base content.
3. AI Summarization. Automated generation of summaries from conversations, articles, or other data within the Platform.
4. AI Agent Assistance. A widget integrated into the User's ticketing system that reads support tickets and suggests relevant knowledge base guides and possible responses to assist human agents.

Stonly may, from time to time, introduce additional AI features. Unless specific supplementary terms are provided for such features, these AI Terms shall apply.

3. Data Privacy and Processing

3.1 Inputs and Outputs

Questions, data, content, or information submitted by the Customer, its Users, or its end-users to Stonly AI ("Inputs"), and answers, suggestions, summaries, or other results generated by Stonly AI based on such Inputs ("Outputs"), are deemed Customer Data under the Terms of Service and are subject to the rights, restrictions, and obligations applicable thereto.
The Customer acknowledges that third parties may independently submit information or materials that generate results identical or similar to Outputs ("Third-Party Results"), and that the Customer has no right, title, or interest in or to any Third-Party Results.

3.2 Third-Party AI Infrastructure

Stonly relies on third-party cloud infrastructure and third-party AI model providers (collectively, "AI Sub-processors") to deliver Stonly AI features. The list of AI Sub-processors is maintained on Stonly’s trust page and updated in accordance with the Data Processing Addendum.
Stonly does not and will not permit AI Sub-processors to use Customer Data to train, fine-tune, or otherwise improve their AI models. Stonly itself does not use Customer Data to train or fine-tune any AI model.

3.3 Data Minimization and Anonymization

Stonly implements technical measures designed to minimize the volume and sensitivity of Customer Data transmitted to AI Sub-processors. These measures may include anonymization or pseudonymization of personal data contained in Inputs prior to processing by external AI models. The Customer acknowledges that while Stonly takes reasonable steps to reduce exposure of personal data, no anonymization technique is infallible.

3.4 Location of Processing

Processing of data related to Stonly AI is hosted in the European Union (France and Sweden) for Customers whose primary data residency is in the European Economic Area, and in the United States for all other Customers. The applicable processing locations for the Customer are set forth in the Data Processing Addendum and on Stonly’s trust page at:
https://trust.stonly.com/

3.5 Data Retention

Stonly stores Inputs and Outputs for the purpose of providing the Services, including reporting and analytics features. Retention of such data is governed by Stonly’s general data retention policy as described in the Data Processing Addendum.

3.6 Processing Instructions

To the extent the Inputs contain personal data (as defined in the Data Processing Addendum), the Customer instructs Stonly to process such personal data for the additional purpose of providing Stonly AI features to the Customer, its Users, and its end-users.

4. Accuracy, Limitations, and Human Oversight

1. Stonly makes no representations, warranties, or guarantees of any kind regarding the quality, accuracy, completeness, or reliability of any Output.
2. The Customer acknowledges and agrees that Outputs are generated by machine learning models and may contain errors, inaccuracies, omissions, or fabricated information ("hallucinations"). Outputs may not reflect the most current or accurate information applicable to the Customer's or its end-users' context or needs.
3. Human oversight. The Customer shall implement appropriate review processes to verify the accuracy and appropriateness of Outputs before relying on them or making them available to end-users in contexts where errors could cause material harm. In particular, the Customer shall not use Stonly AI Outputs as the sole basis for decisions that produce legal effects or similarly significant impacts on individuals without appropriate human review.
4. To the extent permitted by applicable law, the Customer assumes full responsibility for, and any reliance placed on, the accuracy, fitness, completeness, or appropriateness of any Output. The Customer hereby waives any and all claims, rights, or remedies it may have against Stonly, its licensors, or assigns, for any loss, damage, or harm of any kind resulting from, or in connection with, its use of, or reliance upon, Stonly AI and its Outputs.
5. In no event shall Stonly be liable for any direct, indirect, incidental, consequential, or punitive damages arising from any inaccuracies, errors, or inadequacy in any Output. This limitation is in addition to, and does not limit, the general limitation of liability set forth in the Terms of Service.

5. Acceptable Use

In addition to the restrictions on use set forth in the Terms of Service, the Customer shall not, and shall cause its Users and end-users not to:

1. Use Stonly AI to provide regulated professional advice (including but not limited to medical, legal, financial, or tax advice) without appropriate professional oversight and disclaimers.
2. Use Stonly AI for automated decision-making that produces legal effects or similarly significant effects on individuals, without implementing meaningful human oversight.
3. Attempt to extract, reverse-engineer, or reconstruct the underlying AI models, their weights, training data, or architecture, whether through systematic prompting, output analysis, or any other technique.
4. Deliberately input content that is unlawful, harmful, abusive, threatening, defamatory, or otherwise objectionable with the intent of generating harmful Outputs.
5. Use Stonly AI in any manner that violates applicable laws or regulations, including but not limited to data protection, consumer protection, and non-discrimination laws.

6. Transparency and Regulatory Compliance

This section addresses transparency obligations that Stonly considers appropriate in light of applicable regulations, including the EU Artificial Intelligence Act (Regulation (EU) 2024/1689).

6.1 Disclosure of AI Use

1. The Customer acknowledges that Stonly AI features are powered by artificial intelligence and machine learning models. The Customer is responsible for ensuring that its end-users are informed, through appropriate means (such as interface labels, disclosures, or terms of use), that they may be interacting with or receiving content generated by an AI system.
2. Where Stonly AI is deployed in a customer-facing chatbot context, the Customer shall ensure that end-users are made aware that they are communicating with an AI-powered system and not a human agent, unless a human agent is actively participating in the conversation.

6.2 Risk Classification

The Customer is responsible for assessing whether its specific use of Stonly AI falls within a risk category under the EU AI Act or any other applicable AI regulation. Stonly provides Stonly AI as a general-purpose tool for knowledge management and customer support. The Customer shall not deploy Stonly AI in any high-risk use case (as defined under the EU AI Act or equivalent regulation) without first notifying Stonly in writing and ensuring that all applicable requirements, including conformity assessments, risk management, and record-keeping obligations, are met.

6.3 Cooperation

Stonly shall, upon reasonable written request, provide the Customer with available information reasonably necessary for the Customer to fulfill its own transparency and regulatory obligations in connection with its use of Stonly AI, including information about the general nature and capabilities of the AI models used, the categories of data processed, and the technical and organizational measures applied.

7. Intellectual Property and Outputs

1. As stated in the Terms of Service, Customer Data is and remains the full and exclusive property of the Customer. Outputs generated by Stonly AI from Customer Data are deemed Customer Data.
2. The Customer acknowledges that: (a) Outputs are generated by AI models and may not qualify for intellectual property protection under applicable law; (b) similar or identical Outputs may be independently generated for other customers or third parties; and (c) Stonly makes no representation that any Output is original, unique, or non-infringing.
3. Nothing in these AI Terms grants the Customer any right, title, or interest in or to the AI models, algorithms, or technology underlying Stonly AI, all of which remain part of Stonly’s Intellectual Property or that of Stonly’s licensors, as applicable.

8. AI-Specific Indemnification

In addition to the indemnification obligations set forth in the Terms of Service, the following provisions apply specifically to Stonly AI:

1. Stonly’s indemnification. Stonly shall defend, indemnify, and hold harmless the Customer from and against any third-party claim arising directly from a failure by Stonly to comply with its obligations under Section 3 (Data Privacy and Processing) of these AI Terms, provided that: (a) the Customer notifies Stonly promptly in writing; (b) the Customer grants Stonly sole control over the defense and settlement of such claim; and (c) the Customer provides reasonable cooperation at Stonly’s expense.
2. Customer’s indemnification. The Customer shall defend, indemnify, and hold harmless Stonly from and against any third-party claim arising from: (a) the Customer’s use of Outputs in violation of applicable law or these AI Terms; (b) the Customer’s failure to implement adequate human oversight as described in Section 4; or (c) the Customer’s deployment of Stonly AI in a high-risk use case without complying with applicable regulatory requirements.
3. Any indemnification under this Section is subject to the limitation of liability set forth in the Terms of Service.

9. Limitation of Liability

Without limiting the generality of the limitation of liability provisions in the Terms of Service, the Customer acknowledges and agrees that:

1. Stonly shall not be liable for any damages arising from the Customer’s reliance on Outputs, including but not limited to decisions made, actions taken, or information communicated based on such Outputs.
2. Stonly shall not be liable for any temporary unavailability, degradation, or modification of Stonly AI features resulting from changes to or by third-party AI model providers, provided that Stonly uses commercially reasonable efforts to restore or replace affected features.
3. The aggregate liability of Stonly under or in connection with these AI Terms shall be subject to the same cap as set forth in the Terms of Service and shall not be cumulative therewith.

10. Modifications

Stonly AI relies on rapidly evolving third-party AI models and infrastructure. Accordingly:

1. Stonly reserves the right to modify, update, or discontinue specific Stonly AI features at any time, provided that Stonly shall use commercially reasonable efforts to notify the Customer of material changes in advance.
2. Stonly may update these AI Terms from time to time. Updated AI Terms shall be posted at stonly.com/ai-terms and become effective upon posting. Continued use of Stonly AI after such posting constitutes acceptance of the updated AI Terms.

11. Governing Law

These AI Terms are governed by the same governing law and dispute resolution provisions as set forth in the Terms of Service.