Microsoft AI Business Professional AB-730 Manage prompts and chats by using Microsoft 365 Copilot: 187 practice questions

12 of the 187 Manage prompts and chats by using Microsoft 365 Copilot questions in the Certsqill Microsoft AI Business Professional AB-730 bank, shown in full below. Each one carries an explanation for every option, not just the correct one — the wrong answers are where the marks go.

Preparing for Microsoft AI Business Professional AB-730? Take the free 5-min readiness check →

1. Choose a model by task complexity: Select TWO prompt choices that are appropriate.

Select two. More than one option is correct — every correct one is ticked below.

1. Choose a model by task complexity, then verify claims against reports. ✓

   Model selection should reflect the task’s complexity, while checking claims against the reports supports reliable sharing.
2. Choose the largest model and accept its answer without checking.

   Model size does not guarantee correctness and cannot replace verification against the reports.
3. Request a polished summary without sources, audience, or output structure.

   This omits the context and constraints needed for a focused, reviewable executive summary.
4. Save the prompt and treat that action as a recurring schedule.

   Saving supports reuse, but it does not itself schedule future executions.
5. State the goal, audience, sources, structure, and review step. ✓

   This supplies the essential task, audience, references, output requirements, and verification expectation.

Specify purpose, audience, sources, constraints, and format; select a model for task complexity and verify the result.

2. Reference the approved supplier documents for January: Which resource-selection action is most appropriate?

1. Use supplier names only.

   Names do not provide evidence or establish the requested January-through-March comparison scope.
2. Reference every supplier document available, regardless of date or relevance.

   Irrelevant or out-of-period material can distract Copilot and weaken the defined comparison.
3. Reference the approved supplier documents for January through March. ✓

   This selects relevant, authoritative documents and limits the comparison to the requested period.
4. Use a shorter personal note instead of the approved documents.

   A personal note may be incomplete and is not the approved authoritative source requested for the comparison.

Reference approved, relevant supplier documents and limit them to January through March.

3. Save the prompt for later use: Select TWO actions that preserve the prompt for reuse without scheduling it or

Select two. More than one option is correct — every correct one is ticked below.

1. Schedule the prompt to run each Monday.

   Scheduling creates recurring execution, which the scenario explicitly excludes.
2. Grant colleagues access to every referenced sales-plan file.

   Changing file permissions is outside the requested action and is not required merely to preserve a prompt.
3. Save the generated briefing as if it were the reusable prompt.

   A generated result is not the same as saving the instructions for reuse.
4. Save the prompt for later use. ✓

   Saving preserves the prompt instructions for reuse without scheduling execution or changing source permissions.
5. Share the saved prompt instructions with intended colleagues. ✓

   Sharing passes reusable instructions to colleagues; it does not schedule the prompt or grant access to referenced files.

Save the prompt and share its reusable instructions; neither action schedules it or changes permissions.

4. Schedule the prompt for Fridays at 3: Which action meets her requirement?

1. Share the agenda prompt with meeting participants before each Friday.

   Sharing distributes reusable instructions, but recipients must run them and sharing creates no recurring execution.
2. Place the prompt in a Notebook containing the meeting agenda.

   Notebooks organize focused resources and context; they do not themselves establish scheduled prompt runs.
3. Schedule the prompt for Fridays at 3:00 p.m., selecting the weekly recurrence. ✓

   Scheduling establishes the stated run time and recurrence, unlike merely saving or sharing the prompt.
4. Save the agenda prompt with a weekly title and reopen it each Friday.

   Saving and reopening preserves reuse, but Priya would still need to start the prompt manually each Friday.

Use scheduling with the specified time and recurrence; saving, sharing, or organizing alone will not run it.

5. Share the prompt: What sequence should she use?

1. Rename the current chat, then tell the colleague to locate it through chat history.

   Renaming improves organization, but it does not share a reusable prompt or ensure chat access.
2. Copy the knowledge articles into the prompt so sharing automatically grants their access.

   Prompt sharing does not grant access to referenced files; the colleague’s existing permissions remain relevant.
3. Schedule the prompt first, then share the resulting troubleshooting response with the colleague.

   The requirement is reusable instructions, not an automated run or distribution of one retained result.
4. Share the prompt, then ask the colleague to review its instructions and run it with authorized sources. ✓

   Sharing passes reusable instructions; reviewing and running them preserves the colleague’s responsibility for context and source access.

Share the reusable instructions, while relying on separate permissions for referenced support content.

6. Specify the venue’s accessibility requirements and request: What is the most effective refinement?

1. Run the same prompt again and hope the missing requirements appear in the next response.

   Repeating unchanged wording does not address the identified omission or give Copilot the missing planning requirement.
2. Ask for a shorter schedule while keeping the original venue information unchanged.

   Changing length does not correct omitted accessibility requirements or provide relevant planning constraints.
3. Add more broad wording asking Copilot to make the event schedule much better overall.

   Vague extra wording provides no specific accessibility criteria or measurable correction for the response.
4. Specify the venue’s accessibility requirements and request a revised schedule showing how each is met. ✓

   The refinement supplies missing context and a verifiable output criterion directly tied to the response gap.

Correct the identified gap with specific accessibility context and request visible evidence in the revised output.

7. State the current reports and citation requirement in this: What is the smallest effective correction?

1. Add a permanent instruction requiring citations from every future source.

   A broad instruction may help behavior, but it does not identify the current reports needed for this briefing.
2. State the current reports and citation requirement in this prompt. ✓

   Current task-specific sources and requirements guide the briefing; saved preferences or instructions do not replace them.
3. Turn off saved memory and assume the current reports will automatically become authoritative.

   Disabling saved memory does not supply sources or make current reports authoritative for the task.
4. Delete the previous chat so Copilot forgets the preferred concise tone.

   Deleting chat history is distinct from managing saved memory and does not provide current reports or citation instructions.

Provide current sources and requirements directly; memory and instructions personalize behavior but do not replace task context.

8. Search or browse chat history: What should she do?

1. Rename a newly started chat to “Customer kickoff notes” and continue from its empty context.

   Renaming changes organization only; it does not restore the earlier conversation or its unresolved action list.
2. Search or browse chat history, select the descriptively titled conversation, and reopen it. ✓

   History and search help locate the prior conversation; its descriptive title distinguishes it from similar chats.
3. Filter chats by an agent and delete unrelated results until the required conversation remains.

   Filtering narrows the view, while deletion risks removing conversations and is unnecessary for finding the prior chat.
4. Create a Notebook with the customer files and expect the old conversation to appear automatically.

   A Notebook organizes selected resources, but creating one does not automatically locate or reopen a previous chat.

Use chat history or search and select the existing conversation with the descriptive title.

9. Delete the conversation from chat history using its chat: Which action is appropriate?

1. Delete the launch files referenced by the conversation from their source locations.

   Source files are separate objects; deleting them is not the required action for removing the chat conversation.
2. Delete the conversation from chat history using its chat controls. ✓

   Deleting the conversation removes it from chat history, while separate objects and controls may remain unaffected.
3. Rename the conversation to “Archived launch messaging” and leave it in history.

   Renaming changes the title and organization, but the outdated conversation remains in chat history.
4. Turn off saved memory and expect the conversation to disappear from chat history.

   Saved-memory controls and chat-history deletion are distinct, so disabling memory does not remove the conversation.

Delete the conversation itself; renaming, source-file deletion, and memory controls affect different objects.

10. The owner failed to rename the chat with a descriptive: What mistake should the owner diagnose?

1. The owner should delete the chat and recreate it with a more descriptive opening prompt.

   Recreating risks losing the existing conversation; the title can be changed without deleting valid content.
2. The owner failed to rename the chat with a descriptive project handover title after confirming its content. ✓

   Renaming improves identification in history while preserving the conversation’s content and separate source permissions.
3. The owner should move the conversation into a Notebook because chat titles cannot identify projects.

   Chat titles can support organization; Notebooks organize broader resources and are not required to rename this conversation.
4. The owner should share the chat immediately because renaming requires another person’s permission.

   Renaming is an organizational action and does not require sharing or depend on another person’s permissions.

The specific mistake is leaving an uninformative title; rename the existing chat without changing its content.

11. Create a Notebook with selected campaign resources: Which choice best fits?

1. Create a Page containing the campaign brief and selected links for collaborators.

   A Page is an editable shared artifact, not the same as a Notebook that organizes a broader contextual resource collection.
2. Share the campaign prompt and use its referenced files, chats, notes, and links as the team workspace.

   Sharing a prompt passes instructions but does not assemble those resources into a shared project collection or grant access to them.
3. Create a Notebook with selected campaign resources. ✓

   A Notebook organizes selected project resources and focused context for collaboration; permissions for sources and collaborators still apply.
4. Rename the campaign chats.

   Renaming helps identify chats but does not create a shared collection of files, notes, links, and project context.

Use a Notebook to organize selected campaign resources and project context for collaboration.

12. Define the task: Which correction addresses the prompt-design problem rather than merely changing the model?

1. Define the task, audience, context, constraints, output, and suitable model. ✓

   An effective prompt supplies the missing structure and selects a model according to the task’s complexity and needs.
2. Repeat the same request with a larger model.

   Repeating vague wording and increasing model size does not define the desired work or output.
3. Provide every onboarding file and let Copilot infer the task, audience, constraints, and format.

   Unfiltered files do not replace explicit instructions about the task, audience, constraints, or desired output.
4. Choose a smaller model.

   Changing the model alone does not supply the missing task, context, constraints, audience, or output requirements.

Define the task and context, specify the output, and match the model to complexity.

Other Microsoft AI Business Professional AB-730 domains

• Understand generative AI fundamentals — 218 questions →
• Manage business content and collaboration by using Microsoft 365 Copilot — 187 questions →
• Drive business outcomes by using agents — 186 questions →
• All 778 Microsoft AI Business Professional AB-730 questions →
• Microsoft AI Business Professional AB-730 certification: requirements, cost and exam format →