This AI Policy describes how Access to Tools Inc. d/b/a brand.ai ("brand.ai," "we," "us," or "our") develops, deploys, and governs artificial intelligence in our products and services, including the brand.ai platform, APIs, and related services (collectively, the "Services"). It should be read together with our Terms of Service, Acceptable Use Policy, and Privacy Policy. If you or your organization has a separate written agreement with us, that agreement controls to the extent of any conflict with this AI Policy.
How AI Powers the Services
The Services use large language and multimodal AI models to help customers analyze, generate, and govern brand content. Outputs are grounded in each customer’s own brand materials, such as guidelines, strategy documents, and archives, which are supplied to the models as context at the time of each request. Grounding content in this way does not train or modify the underlying models: your brand materials shape your outputs, not the model itself.
No Training on Customer Data
We do not use customer content submitted to the Services, or output generated by the Services, to train or fine-tune any artificial intelligence or machine-learning model, except to the extent we have expressly agreed otherwise in writing with the customer. This commitment applies to brand.ai and flows down to the third-party model providers we use.
Third-Party Model Providers
The Services are powered by foundation models from established providers, currently including Anthropic. Model providers are engaged as subprocessors under written agreements that prohibit the use of customer data to train their models and impose confidentiality and security obligations consistent with our commitments to customers. We review AI subprocessors before use and when their terms or capabilities materially change. A current list of our subprocessors, including AI model providers, is available on our Trust Center.
Human Oversight
The Services are designed to assist and augment human judgment, not replace it. Customers are responsible for reviewing, validating, and approving outputs before publication, distribution, or use in business decisions, as set out in our Acceptable Use Policy. Our brand governance features are aids to compliance, not guarantees of legal or regulatory compliance.
Responsible Use
Our Acceptable Use Policy restricts harmful and high-risk uses of the Services, including:
- Automated decisions with a material impact on individual rights, employment, credit, education, or access to services, without meaningful human oversight
- Deceptive synthetic media, impersonation, or coordinated disinformation
- Biometric identification, social scoring, or exploitation of vulnerable individuals or groups
These restrictions reflect emerging AI regulation, including prohibited-practice and high-risk categories under the EU AI Act, and we enforce them under the Acceptable Use Policy.
Security and Privacy
Customer data processed by the Services is protected by the security program described on our Security page and Trust Center, including encryption in transit and at rest, access controls, and independent audits. Our handling of personal information is described in our Privacy Policy, and we enter into data processing agreements with customers where applicable.
Governance
We maintain an internal AI governance program with executive ownership. New AI features and material changes to our model stack are risk-assessed before release, and we monitor the performance of the Services on an ongoing basis. We maintain compliance with SOC 2 Type II and are pursuing certification to ISO/IEC 42001, the international standard for AI management systems. Current certification status is available on our Trust Center.
Regulatory Developments
AI regulation is developing quickly, including the EU AI Act and U.S. state AI and disclosure laws. We monitor these developments, update this AI Policy and the Services as requirements take effect, and support our customers’ own compliance obligations, including where law requires the disclosure of AI-generated content.
Changes
We may update this AI Policy from time to time by posting the revised version on our website. Material changes will be communicated through the Services or by email.
Contact
If you have questions about this AI Policy, contact us at privacy@brand.ai.
Last updated: September 2026