Skip to content

AI Assistant and Privacy – What You Need to Know

← Security and Privacy

Overview


Diliforce's AI Assistant is powered by generative AI models accessed through Microsoft Foundry. By default, it uses AI models operated within the Microsoft Azure environment. Your organization's administrator can additionally enable optional AI models from another provider — currently Anthropic's Claude models. How your data is handled depends on which model is in use, and both are explained below.

Default AI models (operated by Microsoft)


  • Your prompts and the AI's responses are processed within the Microsoft Azure environment.
  • Data stored at rest stays within the EU geography of the Service. Some processing may take place in other Microsoft regions, including outside the EEA, in which case an appropriate safeguard such as the EU Standard Contractual Clauses applies.
  • Your data is not shared with other customers and is not used to train AI models.
  • The AI provider may process and temporarily retain prompts and responses in accordance with its own terms, for example to detect and prevent misuse.
Optional Claude (Anthropic) models — only if enabled by your administrator

Claude is off by default. An administrator can enable it for your organization in Account Settings.
  • When enabled, Claude runs on Anthropic's own infrastructure, with Anthropic acting as a separate data processor. Microsoft Foundry provides the access, identity and billing layer, while the model processing itself is performed by Anthropic.
  • Because of this, your prompts and responses may be processed outside the EEA. Where data is transferred outside the EEA, an appropriate safeguard such as the EU Standard Contractual Clauses applies.
  • Your data is not shared with other customers and is not used to train Anthropic's models.
  • Anthropic may process and temporarily retain prompts and responses in accordance with its own terms (its Data Processing Addendum and Commercial Terms of Service), for example to detect and prevent misuse.
How Diliforce handles your conversations

Files you add to an AI conversation are stored only for the duration of the interaction and are deleted afterwards. This describes Diliforce's own handling; as noted above, the AI provider may process and temporarily retain prompts and responses under its own terms.

Responsible Use


The AI Assistant supports document-related workflows, but AI-generated content may not always be accurate, complete or up to date. You should review outputs carefully, independently verify important information, and avoid relying on AI-generated responses as legal, financial or other professional advice. Please do not enter special categories of personal data (GDPR Article 9) or health information into the Service. Use of the AI Assistant is subject to Diliforce's applicable service terms.

Further Information