Where the AI summary appears

On your Dashboard, above the Clinic Health Index pillar breakdown, there's a masthead labeled AI summary with a sparkle icon. It shows one or two plain-language sentences describing how your clinic is doing right now.

What you'll learn
  • What data is actually sent to the AI to generate the summary
  • Why the AI can describe your clinic's state but can't state a number
  • What the fallback sentence looks like when the AI summary isn't available
  • Why the "AI summary" label stays visible even when a fallback sentence is shown
AI summary masthead above the Clinic Health Index
The Dashboard with the AI summary masthead visible above the Clinic Health Index pillar breakdown, showing a sparkle icon and a two-sentence summary.

What's actually sent to the AI

The AI summary isn't generated from your clinic's raw data. ClinyPal computes your Clinic Health Index score and band first, entirely outside the AI, using its normal reporting logic. Only that already-computed score, its band, and the titles of your top action items are then sent to the AI.

  • What's sent: the clinic's computed health score and band, plus the titles of top action items.
  • What's never sent: raw counts, revenue figures, appointment numbers, or any other underlying metric.
  • What the AI is instructed to do: write one or two plain-language sentences describing clinic state, without inventing or restating a specific figure.
AI-generated, working from a score, not raw numbers The AI summary is generated by AI, but only from your clinic's already-computed health score, band, and top action-item titles — never from raw numbers. The AI is explicitly instructed never to invent a figure of its own.

When the AI summary isn't available

If the AI call fails or isn't configured for your clinic, the masthead doesn't disappear or show an error. Instead, a canned, non-AI fallback sentence is shown in its place.

The label doesn't always mean AI wrote the sentence The "AI summary" label stays visible on the masthead even when the fallback sentence is being shown. That means the label identifies the feature, not a guarantee that the exact sentence in front of you was generated by AI in that moment — if the AI call failed, you're reading a pre-written fallback under the same label.

FAQ

Was this article helpful?
Your feedback helps us improve our documentation.
Rate this article: