Hi I am getting an issue with using the AI assistant where i keep getting and error about terms not being recognized (shown below) I would appreciate some quick help as I am meant to be demonstrating this to our internal stakeholders today

The error in your screenshot shows that the AI Assistant cannot recognize

fact/Actual_Hours_Booked

as a metric. This is a common issue related to how analytical objects are named and described in your workspace. Here are the most likely causes and quick fixes: Why This Happens The AI Assistant's quality depends heavily on how well your semantic model objects are named and described. Specifically: • The object name

fact/Actual_Hours_Booked

uses a technical naming convention (with a prefix like

fact/

) that the AI Assistant struggles to interpret as a meaningful business term. [AI Best Practices] • The Assistant cannot find semantically similar metrics/facts, which suggests the object lacks a clear, human-readable description or synonyms. [AI Best Practices] Quick Fixes to Try Now 1. Rephrase your prompt using the exact object name — Try referring to it as

Actual Hours Booked

(without the

fact/

prefix) in your question, as the Assistant requires clear, human-readable terms. [Troubleshooting Tips] 2. Reset the chat — Click the reset (rounded arrow) button at the top of the AI Assistant panel and try again with a rephrased, more descriptive prompt. [AI Assistant] 3. Be specific in your prompt — For example, instead of a vague question, try: "What is the total of Actual Hours Booked?" — being specific and detailed gives the agent more context. [Community Slack] Longer-Term Fixes (for after your demo) • Add a clear description to the

fact/Actual_Hours_Booked

object in your workspace, explaining what it measures in plain business language. [AI Best Practices] • Create a metric with a friendly name like

Total Actual Hours Booked

using MAQL, so the AI can easily find and use it. [Create Metrics] • Use AI Memory to map business terms to this fact/metric so the Assistant understands it in future queries. [AI Memory] • Run the Semantic Quality Agent to detect and fix similar naming issues across your workspace. [Semantic Quality Agent] For your demo today, I'd recommend trying option 1 and 2 above as the fastest path to getting a working result. Good luck! 👍 0 button 👎 0 button ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups. 👀 Our Support team monitors all messages and will jump in if needed.

I am getting this issue when directly referencing facts or metrics which last week it was able to do

Hi Chris. I'm sorry to hear that. To solve this, try running the metadata sync. Here is the documentation. https://www.gooddata.com/docs/cloud/api-and-sdk/api/api_reference_all/?apiPath=api-reference%2Fother%23beta-sync-metadata-to-other-services If the issue still persists, please let us know.

That has seemed to work I will do some more testing

Happy to hear that, Chris. Just let us know if you need any further assistance.