Skip to content

Chatting with Insight AI

Open chat from the Chat tab in the dashboard navigation — it opens in its own browser tab so you can keep the dashboard alongside it.

  1. Pick a model from the model selector at the top of the chat (see choosing a model).
  2. Type your message in the input box (“Message Insight AI…”) and press Enter. Use Shift + Enter for a new line.
  3. The conversation is saved automatically and given a short title based on your first message. It appears under Recent in the sidebar.

While a response is being generated, the send button becomes a Stop button so you can cancel mid-response.

If you’re not sure where to begin, the empty chat screen offers suggestion cards such as “Analyze my data pipeline performance” or “Draft a data quality report” — click one to use it as your message.

A new chat: the “What can I help you with?” screen with suggestion cards, the model selector at the top, and the message input below

When you send a message, Insight AI automatically searches all of your workspace’s collections for content related to your question. The most relevant passages — along with the file paths they came from — are provided to the model as context, so answers can be grounded in your organization’s own documents.

Things to know about this behavior:

  • It’s automatic. You don’t select a collection per chat; every ingested document in the workspace is eligible.
  • Only strong matches are used. Passages must be sufficiently similar to your question to be included, and the amount of context is capped, so very loosely related documents won’t influence answers.
  • It’s best-effort. If nothing relevant is found (or your workspace has no ingested documents yet), the model simply answers from its general knowledge.
  • Answers don’t show formal citations. The model may mention the source documents it drew on in its reply, but there is no separate citations panel.

To get grounded answers, make sure your documents have been ingested — see pipelines.

A conversation where the answer draws on an ingested vacation policy document; hovering the response reveals its timestamp, Regenerate, and Copy actions

Hover over any message to reveal its actions:

  • Copy — copies the message text to your clipboard.
  • Edit (your messages) — revise a question and get a fresh answer; see editing & response versions.
  • Regenerate (AI messages) — ask the model to answer again.

AI responses are rendered with full formatting — headings, lists, tables, and syntax-highlighted code blocks.