AI transparency
Eigr is an AI-driven service. You should always know when you are talking to an AI, what it does with your information, and what its limitations are.
| Provider | Model | Used for | Human oversight | Limitations |
|---|---|---|---|---|
| Google (via Lovable AI Gateway) | gemini-2.5-flash / gemini-3-flash | Assistant, contents scanning, document interpretation, translation, summaries | The user confirms every suggestion before it is saved | May make mistakes in recognition, value estimates and document interpretation |
| External chat tools connected by the user (MCP) | The model chosen by the user in that tool (e.g. Claude, ChatGPT) | Reading and updating the user's own Eigr data from an external chat tool | The user approves the connection and can revoke it at any time; access follows the user's own permissions | Eigr does not control the external model; data sent there is processed under that provider's terms |
You are talking to an AI
The assistant in Eigr is an AI. Content generated by AI — summaries, suggestions, value estimates and visualisations — is labelled in the interface.
No automated decisions
The AI makes no decisions with legal effect for you. Every suggestion must be confirmed by you before anything is saved or changed.
Your choices
You can use the product without using the assistant, and you can delete conversations and attachments at any time. The AI provider acts as a processor and does not train models on your content.
Connecting external chat tools
You can connect Eigr to Claude, ChatGPT or other tools that support MCP. The tool signs in with your Eigr account and gets exactly the same access as you — no more. Content you pull into that tool is processed by that tool's provider, not by Eigr. Every action is written to the audit log, and you can revoke the access at any time under Settings.