These AI Terms apply when you use the artificial intelligence features of the Shopxare Services ("AI Features"), including Are AI (the AI agent that answers End Users), Copilot (suggestions for teammates), live translation, mood and urgency detection, knowledge-gap suggestions and AI actions that use data connectors. They are part of the Terms of Service.

1. How AI Features work

AI Features send the information needed for each task — for example the conversation, the help articles and snippets you selected as knowledge, and data returned by connectors you configured — to a large language model and return the result. The model is provided by:

Google (Gemini), when your Workspace uses Shopxare's platform AI; or
the provider whose API key you add (OpenAI, Anthropic or Google), under your own agreement with that provider.

We do not use your Customer Data, your prompts or AI output to train AI models, and we do not allow our platform AI provider to do so.

2. Your inputs and outputs

As between you and us, you own the content you provide to AI Features ("Input") and, to the extent the law allows, the results they produce ("Output"). You are responsible for having the rights to the Input, including the knowledge sources and websites you ask Are AI to learn from.

3. Accuracy and review

AI Output is generated automatically and may be inaccurate, incomplete or out of date, even when it sounds confident. You are responsible for:

the knowledge you give the AI and keeping it current;
the settings that decide when Are AI answers on its own and when it hands over to your team;
reviewing Copilot suggestions and translations before relying on them;
the answers your Workspace sends to End Users, which are treated as messages from you.

Do not use AI Features as the only basis for decisions with legal or similarly significant effects on people (such as credit, employment, insurance, housing, health or legal advice).

4. Transparency to End Users

Are AI identifies itself as an AI agent in the Messenger. You will not remove or disguise that disclosure, and you will not present AI Output to End Users as written by a human. Where the law requires (for example the EU AI Act or state bot-disclosure laws), you will give End Users any additional notice required.

5. Prohibited uses

In addition to the Acceptable Use Policy, you will not use AI Features to:

generate content that is unlawful, harmful, sexually explicit, discriminatory, or that deceives people;
collect or infer sensitive data about End Users, such as health, religion, sexual orientation or political opinions;
identify people through biometrics or monitor them unlawfully;
try to extract the system prompts, model weights or other users' data, or to jailbreak the models;
break the usage policies of the model provider in use.

6. Usage and limits

AI usage is measured in resolutions, requests or credits as described in your Plan. When the included amount is used up, AI Features may pause until the next period or until you buy more credits. We may limit or suspend AI Features to protect the Services or to comply with a provider's policies.

7. Third-party models

AI models are provided by third parties and may change, be updated or be withdrawn by them. We may switch to a comparable model to keep AI Features working. When you use your own API key, your use and costs with that provider are your responsibility.

8. Disclaimer

AI Features are provided "as is". Except as stated in the Terms of Service, we make no warranty that AI Output will be accurate, complete, lawful or suitable for your purpose, and we are not liable for decisions made or actions taken on the basis of AI Output.