Givebutter AI Addendum

1. Overview and Scope.

This AI Addendum (“Addendum”) is effective and binding on Company as of October 1, 2026, and is entered into by Givebutter, Inc. (“Givebutter,” “Company,” “we,” “us,” or “our”), located at 2810 N Church St #53748, Wilmington, Delaware 19802.

This Addendum supplements and is incorporated into the Givebutter Terms of Service (the “Terms”). By accessing or using any AI Features, you agree to the terms of this Addendum.

This Addendum applies to all AI Features, including Homegrown Models, Commercial AI Features, and AI Infrastructure Services (each as defined in Section 2 below).

Capitalized terms used but not defined in this Addendum have the meanings given to them in the Terms.

2. Definitions.

The following terms have the meanings set forth below when used in this Addendum.

1. “Artificial Intelligence” or “AI” means machine-based technology that, for explicit or implicit objectives, infers from inputs how to generate outputs such as predictions, content, recommendations, classifications, decisions, or other results, including through machine learning, deep learning, natural language processing, large language models, generative AI, or other similar or successor technologies. For clarity, Artificial Intelligence does not include rules-based automation or ordinary-course software functionality that does not infer from inputs how to generate outputs.
2. “AI Features” means all AI-powered and machine learning features available through the Services, including Homegrown Models, Commercial AI Features, AI Infrastructure Services and AI Connectors.
3. ‍"AI Connectors” means integrations that allow Fundraisers to query their Givebutter data through external AI platforms using Givebutter’s MCP server.
4. “AI Model Providers” means third-party companies providing AI model services used by Commercial AI Features. As of the Effective Date, these providers are Anthropic, Google, and OpenAI.
5. “Commercial AI Features” means AI Features that transmit data to AI Model Providers to generate outputs.
6. “Homegrown Models” means predictive models built and trained by Givebutter using aggregated platform data that has been de-identified before use in model training, running entirely within Givebutter’s infrastructure.
7. “AI Infrastructure Services” means AI infrastructure and observability tools that log requests to and responses from AI model providers for quality monitoring and performance evaluation.

3. Categories of AI Features.

Givebutter’s AI Features fall into four categories:

1. Homegrown Models: Models built and trained by Givebutter using aggregated platform data that has been de-identified before use in model training. These models run entirely within Givebutter’s infrastructure and do not transmit data to AI Model Providers. Current examples include: (i) Suggested Donation Amounts, which predicts expected donation sizes using historical transaction data, campaign metadata, and account information; and (ii) Next-Action Recommendations, which predicts the next product action a Fundraiser should take based on aggregated and de-identified behavioral usage data.
2. Commercial AI Features: Features that transmit data to AI Model Providers (currently Anthropic, Google, and OpenAI) to generate outputs. Current examples include Campaign Creator, Contact Summaries, Smart Filters, Custom Reports, and Receipt Matching. The specific data transmitted depends on the feature and may include personal information, as described in Section 4.
3. AI Infrastructure Services: AI infrastructure and observability tools that log requests to and responses from AI Model Providers for quality monitoring and performance evaluation. AI Infrastructure Services include forwarding and proxy service applications to route requests to providers. Logs may contain personal information where applicable. We will take reasonable steps to seek contractual commitments from these service providers to restrict the use of your data for AI model training; however, in some cases, data captured by such services may be used for AI model training.
4. AI Connectors: Integrations that allow Fundraisers to query their Givebutter data through external AI platforms using Givebutter’s MCP server. When enabled, the external AI platform may access data from the Fundraiser’s Givebutter account, excluding banking data.

4. Data Practices for AI Features.

4.1 Homegrown Models. Training data for Homegrown Models is derived from aggregated platform data that has been de-identified before use in model training. No donor names, emails, payment credentials, or other direct identifiers or personal information are used as model features. Join keys (such as account or campaign IDs) are used only to assemble training datasets and are not passed to, or used by, the model.

4.2 Commercial AI Features. Data is transmitted to the applicable AI Model Provider only as needed to generate the requested output. The specific data transmitted depends on the feature and the user input and may include personal information.

4.3 AI Infrastructure Services. All AI prompts and responses pass through an AI infrastructure and observability service (currently Braintrust) that logs requests and responses for quality monitoring, performance evaluation, and debugging. These logs may contain the same data transmitted to the AI Model Provider, including personal information where applicable.

4.4 Additional Consent. By using the Services, you consent to Company retrieving your logo, brand colors, and other publicly available brand elements from your website to personalize your experience on the Site and within the Services.

A Fundraiser’s identifiable data is not shared with other Fundraisers through AI features.

5. Third-Party AI Model Providers.

5.1 Givebutter’s commercial agreements with its AI Model Providers are designed to prohibit those providers from using data submitted through the Services for training or improving their models.

5.2 AI Model Provider processing is designed to be ephemeral: inputs are generally processed to generate the requested output and are not retained by the provider beyond the time needed for that purpose, subject to the provider’s own retention policies.

5.3 A current list of Givebutter’s subprocessors, including AI Model Providers, is available through our Trust Center. We will update this list as such providers change.

6. AI Connectors and Third-Party AI Platform Integrations.

6.1 By enabling an AI Connector, the Fundraiser consents to and directs Givebutter to make its data available to the external AI platform selected by the Fundraiser, and represents and warrants that it has obtained all necessary consents and authorizations to direct such sharing. Givebutter acts on the Fundraiser’s instruction and does not control how the external platform processes that data.

6.2 Givebutter is not responsible for the privacy, security, or data handling practices of any external AI platform accessed through an AI Connector. The Fundraiser is responsible for reviewing and accepting the terms, privacy policies, and data practices of any external AI platform it connects via an AI Connector.

6.3 Data transmitted to external AI platforms through AI Connectors will be subject to those platforms’ own data use and retention policies, including policies that permit the use of submitted data for model training. Certain consumer-tier accounts with these platforms may permit training on user data. The Fundraiser acknowledges and understands that enabling an AI Connector may expose its data to such training practices and should review the applicable platform’s terms before enabling an AI Connector.

6.4 The Fundraiser’s use of AI Connectors is at its own risk. The disclaimers and limitations of liability set forth in the Terms apply fully to the Fundraiser’s use of AI Connectors and to any outputs generated by external AI platforms using the Fundraiser’s data.

7. AI-Generated Content: Disclaimers and Responsibilities.

7.1 AI-generated Content is provided as a suggestion only and is not a substitute for the User’s independent judgment. It may contain errors, omissions, inaccuracies or hallucinations. AI-generated content does not constitute professional, legal, financial, tax, compliance, fundraising, or other advice, and should not be relied on as such.

7.2 Users are solely responsible for reviewing, verifying, and approving any AI-generated Content before publishing, sending, relying on, or otherwise using it. This includes, without limitation, content generated by the Campaign Creator feature.

7.3 Givebutter does not guarantee the accuracy, completeness, reliability, or suitability of AI-generated Content for any particular purpose. To the fullest extent permitted by applicable law, AI-generated Content is provided “AS IS” and without any warranty, express or implied. The disclaimers and limitations of liability in the Terms apply to all AI-generated Content.

7.4 Content generated using a User’s data belongs to such User, subject to Givebutter’s underlying intellectual property rights in the platform, models, algorithms, and AI infrastructure. By using AI features, the User grants Givebutter a non-exclusive, royalty-free license to use, reproduce, and display such content solely as necessary to provide and improve the Services. Givebutter may generate similar or identical content for other users and retains all rights to use aggregated or anonymized data derived from AI feature usage.

8. Opt-Out and Controls.

8.1. AI Features are enabled by default as part of the Services.

8.2. Fundraisers may opt out of certain AI Features on a feature-by-feature basis by using the in-product controls. Not all AI Features support individual opt-out; certain features that are integral to core platform functionality (such as AI-assisted fraud detection and Receipt Matching) cannot be disabled.

8.3. Fundraisers can opt out of their data being used for Homegrown Model training by contacting Givebutter Support.

8.4. AI Connectors require affirmative opt-in. Fundraisers enable AI Connectors by activating the applicable integration and may disable the integration at any time through the third-party provider.

9. Homegrown Model Training.

9.1. Givebutter uses aggregated data derived from platform activity to train and improve its Homegrown Models. Before use in model training, this data is stripped of all direct identifiers and personal information, including names, email addresses, phone numbers, physical addresses, and payment credentials.

9.2. Training data may be derived from activity across the platform, including activity from multiple Fundraisers and Supporters.

9.3. Givebutter does not currently fine-tune or post-train third-party AI models using customer data. If Givebutter elects to do so in the future, we will update this AI Addendum before any such processing begins, and obtain any required consents.

10. Changes to AI Features.

Givebutter may add, modify, or discontinue AI Features at any time in its sole discretion. We will update this AI Addendum to reflect material changes to AI Features and will provide notice of material changes in accordance with the Terms.

Without limiting the foregoing, the following types of changes will be reflected in updates to this AI Addendum and, where applicable, the Privacy Policy and our subprocessor list:

1. Changes to the AI Model Providers we use.
2. Changes to the categories of data processed by AI Features.
3. Introduction of new AI Features that materially expand the scope of data processing beyond what is described in this AI Addendum.

Continued use of the Services after any such update constitutes acceptance of the revised AI Addendum. Fundraisers that do not agree to a revised Addendum should discontinue use of the affected AI Features or, if applicable, the Services.

11. Relationship to Other Governing Documents.

11.1 This AI Addendum is incorporated into and subject to the Terms. In the event of a conflict between this Addendum and the Terms, this Addendum controls with respect to AI-related matters.

11.2 Data processing practices for AI Features are further described in the Privacy Policy. The Privacy Policy supplements this Addendum with respect to the collection, use, and disclosure of personal information in connection with AI Features.

11.3 For Fundraisers that are subject to the Data Processing Addendum, the DPA governs Givebutter’s processing of personal data that the Fundraiser uploads or imports or which Givebutter accesses to provide the Services and for which the Fundraiser is the controller. To the extent an AI Feature processes data governed by the DPA, the DPA’s terms apply to that processing in addition to this AI Addendum.

11.4 Except as expressly modified by this AI Addendum, all terms and conditions of the Terms, Privacy Policy, and DPA remain in full force and effect.