Foundation Model Supplementary Terms of Service

1. Incorporation and Definitions

These Supplementary Terms form an integral part of the Master Services Agreement or Personal Use Terms of Service (“Agreement”) between you and UtopikAI. They govern your use of features powered by Foundation Model Providers (“FMPs”), including Large Language Models (LLMs) and content generation models.

All capitalized terms have the meaning assigned in the Agreement.

2. Service Evolution and Modifications

• Foundation Model services are evolving. UtopikAI reserves the right to modify these terms in response to changes introduced by FMPs.
• UtopikAI is not liable for impacts resulting from FMP policy or functionality changes.

3. Data Processing and Privacy

a. Customer Data Transmission

You authorize UtopikAI to transmit Customer Data, including Customer Personal Data, to Foundation Model Providers in accordance with:

• UtopikAI’s Data Processing Agreement (DPA)
• Functional requirements of the service

FMP-related DPA agreements are listed in the Sub-Processors list.

b. Model Training Restrictions

• FMPs (OpenAI, Anthropic, Mistral, Google, Fireworks.ai) must not use Customer Data for training
• UtopikAI must not authorize third parties to use Customer Data for training
• Your usage of UtopikAI’s Services does not grant any model training rights

c. Zero Data Retention

• FMPs (OpenAI, Anthropic, Mistral, Google, Fireworks.ai) follow a Zero Data Retention policy
• Customer content is not logged, stored, or reviewed
• Certificates are available in the Sub-Processors list

d. Model Provider Selection

Customers can:

• Opt out of specific FMPs used for embeddings (upon request)
• Select FMPs for inference directly in the user interface

4. Input and Output Provisions

a. Definitions

• Input: content submitted for processing
• Output / Foundation Model Output: content generated by an FMP
• Both are considered User Content under the Agreement

b. User Responsibilities

You are solely responsible for:

• Legal compliance in your use of Inputs and Outputs
• Adhering to these terms, the Master Agreement, and Acceptable Use Policy
• Respecting third-party intellectual property rights

5. Prohibited Activities

You are strictly prohibited from:

• Developing competing foundation models or services
• Using Outputs to mislead or deceive
• Generating manipulative content (e.g. spam, electoral materials)
• Violating UtopikAI’s technical documentation or usage limits

6. Third-Party Compliance

You must comply with all applicable terms and policies from the Foundation Model Providers, including:

• OpenAI
• Anthropic
• Mistral AI
• Google Generative AI APIs
• Fireworks.ai

7. Warranty Disclaimer

a. UtopikAI Disclaims All Warranties Regarding:

• Results obtained through the use of Services
• Accuracy or reliability of Foundation Model Outputs

b. Risk Acknowledgment

• You use the Services at your own risk
• You must independently verify Output accuracy
• Outputs may contain inaccuracies or become outdated
• No oral or written advice from UtopikAI constitutes a warranty

‍