A is incorrect: Screenshots capture visual snapshots but cannot be used as Copilot references for grounding future responses. The Add to Notebook option preserves conversation content in a format that Copilot can read, analyze, and use as context. Reference: https://support.microsoft.com/en-us/topic/revisit-your-microsoft-365-copilot-chat-history-6ea899e3-3bb1-450a-a2ae-220341ac193a

B is incorrect: Manually retyping a conversation into a Copilot Page is error-prone and time-consuming. The Add to Notebook option in chat history directly adds the conversation as a notebook reference with a single action, preserving the original content. Reference: https://support.microsoft.com/en-us/topic/revisit-your-microsoft-365-copilot-chat-history-6ea899e3-3bb1-450a-a2ae-220341ac193a

C is incorrect: Manually copying and pasting Copilot responses into SharePoint does not integrate the conversation into a Copilot Notebook as a reference. The Add to Notebook option preserves the full conversation and makes it directly available for grounded context. Reference: https://support.microsoft.com/en-us/topic/revisit-your-microsoft-365-copilot-chat-history-6ea899e3-3bb1-450a-a2ae-220341ac193a

D is correct: The Add to Notebook option in the conversation's More menu adds the chat conversation directly to a Copilot Notebook as a reference. This preserves the conversation content and lets Copilot use it as grounded context for future prompts within that notebook. Reference: https://support.microsoft.com/en-us/topic/revisit-your-microsoft-365-copilot-chat-history-6ea899e3-3bb1-450a-a2ae-220341ac193a

A is incorrect: Copilot does not bypass label protection by adding a watermark. If the teammate lacks the necessary usage rights, Copilot will not process the encrypted content to produce a summary in the first place. Reference: https://learn.microsoft.com/en-us/copilot/microsoft-365/microsoft-365-copilot-architecture-data-protection-auditing

B is correct: Copilot honors Microsoft Purview sensitivity labels and encryption. When content is encrypted, the user must have the appropriate usage rights for Copilot to interact with it, so a teammate without those rights cannot have Copilot summarize the protected document. Reference: https://learn.microsoft.com/en-us/copilot/microsoft-365/microsoft-365-copilot-architecture-data-protection-auditing

C is incorrect: Copilot does not downgrade labels to fit a user's access. It respects the existing label and protection settings, preventing users without the required rights from having the protected content summarized. Reference: https://learn.microsoft.com/en-us/microsoft-365/copilot/enterprise-data-protection

D is incorrect: Copilot does not strip sensitivity labels to enable broader sharing. It enforces the label's protection, and a user lacking the required usage rights cannot have the content summarized at all. Reference: https://learn.microsoft.com/en-us/microsoft-365/copilot/enterprise-data-protection

A is incorrect: Informally shared personal prompts will not produce a consistent, governed experience across 30 people. A single agent with tailored suggested prompts and instructions is the scalable solution. Reference: https://learn.microsoft.com/en-us/microsoft-365/copilot/extensibility/agent-builder

B is correct: When a unique business process requires standardized behavior and starting points for many people, a new agent with tailored instructions and suggested prompts delivers a consistent, guided experience that general chat cannot guarantee. Reference: https://learn.microsoft.com/en-us/microsoft-365/copilot/extensibility/agent-builder-build-agents

C is incorrect: A generic agent is not tailored to your unique onboarding process, instructions, or terminology. The requirement for process-specific guidance justifies building a new agent. Reference: https://learn.microsoft.com/en-us/microsoft-365/copilot/extensibility/agents-overview

D is incorrect: Relying on individuals to remember phrasing produces exactly the inconsistency you are trying to solve. An agent with suggested prompts standardizes the experience automatically. Reference: https://learn.microsoft.com/en-us/microsoft-365/copilot/extensibility/agents-overview

A is incorrect: Copilot in Teams summarizes Teams meetings and chats, not Outlook email threads. The thread should be summarized with Copilot in Outlook, where the messages reside. Reference: https://support.microsoft.com/en-us/office/summarize-an-email-thread-with-copilot-in-outlook

B is incorrect: Moving an email thread into a channel adds unnecessary steps and may expose sensitive pricing details. Copilot in Outlook can summarize the thread directly where it already exists. Reference: https://support.microsoft.com/en-us/office/summarize-an-email-thread-with-copilot-in-outlook

C is incorrect: Your director wants a summary of this specific negotiation, not generic market information. Copilot in Outlook should summarize the actual email thread, not external web content. Reference: https://support.microsoft.com/en-us/office/summarize-an-email-thread-with-copilot-in-outlook

D is correct: The negotiation occurred entirely within an Outlook email thread, so Copilot in Outlook is the right tool to condense the multi-party exchange into a short summary of the outcome for your director. Reference: https://support.microsoft.com/en-us/office/summarize-an-email-thread-with-copilot-in-outlook

A is incorrect: The semantic index helps Copilot find relevant documents but does not validate whether the AI accurately represents the content of those documents. Both internal and web citations require human review because Copilot may include information not present in source materials. Reference: https://learn.microsoft.com/en-us/microsoftsearch/semantic-index-for-copilot

B is incorrect: The number of citations does not indicate accuracy. Copilot can provide citations while still including ungrounded information. The correct approach is to verify that each citation actually supports the claim it references and to identify any unsupported statements. Reference: https://learn.microsoft.com/en-us/copilot/microsoft-365/microsoft-365-copilot-transparency-note

C is correct: Users should verify AI-generated content by clicking citations to confirm the source material supports the claims, and identify uncited statements that may be ungrounded. Both organizational and web citations require verification because Copilot may misrepresent source content or generate unsupported claims. Reference: https://learn.microsoft.com/en-us/copilot/microsoft-365/microsoft-365-copilot-transparency-note

D is incorrect: The Microsoft 365 admin center does not include an automated citation accuracy scanning tool. Citation verification is a human responsibility. Users should click each citation to verify source alignment and review for uncited claims before sharing content externally. Reference: https://learn.microsoft.com/en-us/copilot/microsoft-365/microsoft-365-copilot-transparency-note

A is incorrect: Commissioning a study to design multiple custom agents is excessive for a standard task already served by a maintained Agent Store agent. The simplest fit-for-purpose choice is to add the existing ready-made agent. Reference: https://learn.microsoft.com/en-us/microsoft-365/copilot/copilot-agent-store

B is incorrect: Recreating functionality that an existing maintained agent already delivers wastes effort, especially for someone without building experience. Adopting the ready-made Agent Store agent is the appropriate choice here. Reference: https://learn.microsoft.com/en-us/microsoft-365/copilot/copilot-agent-store

C is incorrect: Using a ready-made agent from the Agent Store requires no development skills. There is no need to delay adoption to learn building techniques when a maintained agent is already available. Reference: https://learn.microsoft.com/en-us/microsoft-365/copilot/copilot-agent-store

D is correct: For a standard, widely adopted task where a maintained agent already exists, adopting it from the Agent Store is the fastest, lowest-effort option, especially for a user without building experience. Reference: https://learn.microsoft.com/en-us/microsoft-365/copilot/copilot-agent-store

A is incorrect: A saved prompt speeds up reuse but still requires you to start it manually each month. Because the cadence is predictable and you want it to run automatically, scheduling the prompt is the better fit. Reference: https://learn.microsoft.com/en-us/copilot/microsoft-365/save-and-reuse-prompts

B is incorrect: Pinning a conversation only makes it easier to locate; it does not run automatically on a schedule. The Schedule a prompt feature is designed to run recurring prompts without manual effort. Reference: https://learn.microsoft.com/en-us/copilot/microsoft-365/find-previous-conversations

C is incorrect: A calendar appointment only reminds you to act manually; it does not run the Copilot prompt. Scheduling the prompt within Copilot automates the recurring monthly summary directly. Reference: https://learn.microsoft.com/en-us/copilot/microsoft-365/copilot-scheduled-prompts

D is correct: Scheduling a prompt lets you set a recurring cadence, including monthly, so the budget email summary runs automatically at the time you choose. This matches a predictable monthly close workflow without requiring you to trigger it each time. Reference: https://learn.microsoft.com/en-us/copilot/microsoft-365/copilot-scheduled-prompts

A is incorrect: Renaming the chat makes it easier to find but does not give colleagues a shared space to edit the procedure together. Use a Copilot Page for collaborative authoring. Reference: https://learn.microsoft.com/en-us/copilot/microsoft-365/manage-conversations

B is incorrect: A notebook gathers related content for grounding and reference but is not built for several people to co-edit one document simultaneously. A Copilot Page supports shared editing. Reference: https://learn.microsoft.com/en-us/copilot/microsoft-365/copilot-notebooks

C is correct: A Copilot Page provides a single shared document where multiple colleagues can co-edit the procedure and keep refining it with Copilot, making it ideal for a living document maintained by a distributed team. Reference: https://learn.microsoft.com/en-us/copilot/microsoft-365/copilot-pages

D is incorrect: Generating separate versions produces inconsistent copies that must be merged later. A Copilot Page lets the team maintain one shared, continually updated document. Reference: https://learn.microsoft.com/en-us/copilot/microsoft-365/copilot-pages

A is incorrect: Copilot does not block deletion based on keywords in a conversation. Compliance preservation is handled through retention and eDiscovery policies, not by preventing you from deleting your own chat. Reference: https://learn.microsoft.com/en-us/purview/retention-policies-copilot

B is incorrect: Deletion clears the conversation from your history, but content governed by retention or legal hold can still be preserved by your organization. Personal deletion does not guarantee permanent destruction. Reference: https://learn.microsoft.com/en-us/purview/retention-policies-copilot

C is correct: Deleting a chat from your personal history does not override organizational governance. Retention policies and eDiscovery in Microsoft Purview can preserve Copilot interactions for compliance, so deletion is not a guaranteed way to permanently destroy records. Reference: https://learn.microsoft.com/en-us/purview/retention-policies-copilot

D is incorrect: Deletion does not transfer ownership of a conversation to anyone. Whether content is retained depends on organizational retention and eDiscovery policies that operate independently of your deletion action. Reference: https://learn.microsoft.com/en-us/purview/ediscovery-search-and-delete-copilot-data

A is incorrect: This prompt lacks specifics about the target role, desired sections, document length, and tone. Without these details, Copilot generates generic content that requires significant revision to match your actual needs. Reference: https://support.microsoft.com/en-us/office/draft-and-add-content-with-copilot-in-word-069c91f0-9e42-4c9a-bbce-fddf5d581541

B is incorrect: While detailed, this prompt is overly broad in scope. Requesting content for all roles and departments produces an unfocused draft. Narrowing the audience and specifying clear sections yields more targeted and actionable output from Copilot. Reference: https://support.microsoft.com/en-us/office/draft-and-add-content-with-copilot-in-word-069c91f0-9e42-4c9a-bbce-fddf5d581541

C is incorrect: Delegating all content decisions to Copilot without specifying audience, sections, or tone produces unpredictable and generic results. Providing clear and specific instructions gives Copilot the context needed to generate a useful first draft. Reference: https://support.microsoft.com/en-us/office/draft-and-add-content-with-copilot-in-word-069c91f0-9e42-4c9a-bbce-fddf5d581541

D is correct: This prompt specifies the target audience (software engineers), scope (3 pages), structure (specific sections), and tone (professional). Microsoft recommends being as detailed as possible when drafting, as more information produces more helpful and specific Copilot output. Reference: https://support.microsoft.com/en-us/office/draft-and-add-content-with-copilot-in-word-069c91f0-9e42-4c9a-bbce-fddf5d581541