Generative AI Source Grounding
Tie each answer to a supplied source rather than a plausible completion.
How it works
Grounding means using relevant evidence to support an answer. Supplying a document can help, but the model may still misread it or add unrelated knowledge. Define what counts as support and what to do when the source is silent. For a bounded document question, a short exact supporting passage can be more useful to a reviewer than a confident explanation without a traceable basis.
How to use this lesson
This is an authored prompt exercise with a human-written acceptance guide. It is not a captured provider response or a promise that a model will produce the target. No provider account, API, model weights or live inference was used for the course execution review. If you try it in an external service, use an authorised environment, check its actual access and data settings, and record the product/model and run date.
Bhylu does not run a generative model or send this prompt to an AI service. Copy and download preserve the text only. External tools may require accounts, paid access or compatible inputs; this course does not promise availability or prices.
Prompt exercise
Source W1: The collection desk closes at 16:30.
Question: When does the desk close?
Return the answer and the source ID. Use no other source.
Acceptance guide — not a model transcript
Reference: 16:30, supported by W1.
The question is answerable from one sentence. A citation validator can confirm that W1 was supplied, but understanding whether a complex claim follows from it still needs a semantic check. Treat source ID validity and actual claim support as two different evaluation dimensions.