AB-730 AI Business Professional Question and Answers

Prompt injection is a generative AI security risk where an attacker inserts instructions (often hidden in text, documents, webpages, or user inputs) to override or manipulate the assistant’s intended behavior. This can lead to unintended actions such as ignoring policy controls, producing unsafe outputs, or attempting to reveal sensitive information. Because generative AI systems follow natural-language instructions, they can be socially engineered to prioritize malicious content unless safeguards are in place. This is why prompt injection can cause data exposure (for example, attempting to extract confidential content from grounded sources) and can also embed harmful instructions that redirect the model’s behavior. In enterprise settings like Microsoft 365 Copilot, mitigations include grounding boundaries, permission trimming, content filtering, and instruction hierarchy (system policies over user instructions). From a business governance perspective, users should treat untrusted inputs (emails, documents, web text) as potentially hostile and apply least-privilege access and validation when using AI outputs in decision-making.

Microsoft 365 Copilot in Teams is designed to provide real-time meeting assistance, including summarizing discussions, identifying key decisions, and highlighting action items. When joining a meeting late, the most efficient and immediate method to catch up is to directly prompt Copilot within the Teams meeting experience.

Option A is the best answer because Copilot in Teams can summarize what has occurred so far during the live meeting. It leverages the meeting transcript, speaker attribution, and contextual signals to generate a concise, structured summary in real time. This minimizes delay and provides actionable insights instantly.

Reading the transcript (Option B) would require manual review and is not the fastest method. Intelligent recap (Option D) is typically available after the meeting concludes and processes recording data. Asking Copilot from the separate Microsoft 365 Copilot app (Option C) may not provide immediate, in-meeting contextual awareness.

Therefore, the fastest and most effective approach is to ask Copilot in Teams to summarize what you missed.

When transcription is not enabled , Microsoft 365 Copilot in Teams cannot access a full historical record of the meeting conversation. Copilot relies on transcript data to provide complete summaries and recaps. Without transcription, Copilot can still assist during the meeting , but only from the point at which it is activated.

Copilot does not automatically activate at the start of the meeting; a user must explicitly invoke it. Once activated, Copilot processes the live conversation stream and can generate summaries, key points, or insights from that moment forward only . It cannot reconstruct earlier discussion because no transcript exists for prior content.

Options B and D are incorrect because they assume automatic activation and full-meeting recap capability, which require transcription. Option A is incorrect because Copilot summarizes the meeting content, not just the speech of the activating user.

Therefore, the correct answer is C .

==================================================

Microsoft 365 Copilot in Teams relies heavily on meeting transcripts to provide post-meeting intelligence such as summaries, key points, decisions, and action items. The transcript serves as the structured textual data that Copilot analyzes to understand what was discussed during the meeting.

The feature commonly referred to as recap or intelligent recap is generated based on this transcript. Without transcription enabled, Copilot lacks the necessary detailed conversation data to accurately reconstruct the meeting content. As a result, generating a recap of what was discussed is only possible when a transcript is available.

Other options such as scheduling a follow-up meeting or identifying attached content do not depend on transcription, as they rely on calendar data or meeting metadata rather than conversational analysis. However, summarization and recap capabilities are directly tied to the transcript.

Therefore, the correct completion is generate a recap of what was discussed during the meeting , as this action specifically requires the meeting to be transcribed.

Microsoft 365 Copilot is designed to be helpful by using work context —for example, the files you have access to, recent activity, meetings, emails, and SharePoint/OneDrive content—to suggest relevant prompts and help you start tasks faster. It also uses this context to augment your prompt before it is sent to the LLM. This is the grounding approach (often described as retrieval-augmented generation): Copilot retrieves relevant organizational content you’re permitted to access and adds it as supporting context so responses are accurate and business-relevant. However, Microsoft 365 Copilot does not use your organization’s contextual data to train the underlying foundation model . That separation is critical for enterprise privacy and compliance: your prompts, responses, and tenant data are used to generate the answer for your session and permissions, but are not used to improve or retrain the base LLM. This approach supports responsible AI, protects confidential business information, and ensures outputs respect access controls.

suggested prompts

In Microsoft 365 Copilot, suggested prompts are designed to help users get started quickly and interact with an agent effectively. Microsoft guidance for Copilot agents emphasizes that many users struggle most at the beginning—knowing what to ask, how to phrase requests, and how to take advantage of an agent’s purpose. Suggested prompts address this by providing ready-to-use examples that demonstrate the best ways to engage the agent for its intended scenarios.

Suggested prompts also act as lightweight “guardrails” for good prompting. They show users the expected input patterns, the types of tasks the agent can perform, and the kinds of outputs they can request. This accelerates adoption and improves the quality of results by steering users toward well-structured requests and appropriate context.

By contrast, capabilities describe what an agent can do (tools/features), instructions define how the agent should behave, and agent settings control configuration. None of those are primarily meant to help users begin conversations through example starters.

Therefore, the correct completion is suggested prompts .

Microsoft distinguishes Microsoft 365 Copilot (the full, licensed experience integrated across Microsoft 365 apps and Microsoft Graph) from Microsoft 365 Copilot Chat (the chat-first experience available more broadly). Some advanced capabilities—especially those tied to specialized agents and persistent work organization—are typically associated with the full Copilot license.

The Researcher agent is an advanced capability and is not consistently available in Copilot Chat across tenants because it is treated as a premium agent experience. Copilot Notebooks are also a structured organization feature used to group conversations and share reference materials; this is typically positioned as part of the full Microsoft 365 Copilot experience.

Copilot Pages , however, is the collaborative, editable canvas used to refine, organize, and share Copilot-generated content. Pages is designed to work alongside Copilot Chat so users can move from a chat response into an editable artifact for collaboration and reuse. This makes Pages the feature most commonly available in both the full Microsoft 365 Copilot and Microsoft 365 Copilot Chat experiences.

Therefore, the best answer is Copilot Pages .

All three statements are false because generative AI responses are not guaranteed to be identical even when the prompt is the same. First, when grounded in the web , results can vary due to changing web content, different retrieved sources, or differences in how information is summarized at run time. Second, when grounded in your organization’s data , responses can change based on updates to files, emails, meetings, permissions, or which specific items Copilot retrieves as the most relevant context at that moment. Third, even when relying only on the model’s general knowledge, large language models are probabilistic: they may choose different wording, structure, examples, or emphasis across runs, especially when temperature/decoding settings and internal routing differ. In business scenarios, this means Copilot outputs should be treated as drafts that may require validation, and repeatability should be improved by adding precise constraints (cite specific sources, use fixed formats, specify exact sections, and request verbatim quotes where appropriate).

To generate an effective monthly summary with Copilot in Excel, Microsoft prompting guidance emphasizes two core elements: clear intent and expected output structure . First, you should state the goal explicitly—such as “summarize total sales by month based on OrderDate and Amount, and highlight month-over-month changes.” This directs Copilot to the correct aggregation and analysis behavior (Option D).

Second, specifying the desired format improves usability and reduces ambiguity. For example, requesting “a table with Month, Total Sales, Number of Orders, and Average Order Value” or “a bulleted executive summary plus a small summary table” ensures Copilot returns results in a format you can immediately use (Option A).

A specific cell range (Option B) is usually unnecessary when the data is already in a recognized Excel table, because Copilot can work from the table context. The workbook file name (Option C) does not help analysis. Providing a formula (Option E) is not required because Copilot can compute monthly totals; formulas are optional if you want a specific calculation method, but they are not the best prompt ingredients for an effective summary.

Therefore, the best answers are A and D .

Microsoft 365 Copilot follows Microsoft’s Responsible AI principles and enforces strict content safety policies. When a prompt violates safety guidelines—such as containing harmful, abusive, illegal, or restricted content—the system may refuse to generate a response. The refusal message shown is consistent with safety filtering behavior.

Generative AI systems include moderation layers that evaluate prompts before generating output. If the prompt is classified as unsafe or non-compliant with policy, Copilot blocks the request and encourages the user to try a different topic.

A vague prompt typically results in a generic or clarifying response rather than a refusal. There is no fixed limit of five requests per prompt. Exceeding the context window usually results in truncation or processing errors, not a safety-based refusal message.

Therefore, the most likely cause of the response is that the prompt contains language that violates safety guidelines.

==================================================

Microsoft provides centralized activity management controls through the My Account portal, which allows users to manage privacy settings, activity history, and data associated with Microsoft 365 services, including Copilot. When the requirement is to delete all conversations with minimal effort, the most efficient method is to use the account-level activity management tools rather than deleting conversations individually.

The My Account portal enables bulk management of Copilot activity data, allowing users to clear conversation history in a consolidated manner. This approach aligns with Microsoft’s privacy-by-design framework, giving users control over their AI-generated interaction history without requiring administrative intervention.

Using the Copilot web or desktop app would typically require manually deleting conversations one at a time, increasing effort. The Windows 11 Settings app is unrelated to Microsoft 365 Copilot data management.

Therefore, to delete all Copilot conversations efficiently and with the least amount of effort, the correct approach is to use the My Account portal in Microsoft 365.

Microsoft 365 Copilot fully respects Microsoft Purview Information Protection, including sensitivity labels applied to documents. When content is used as a source to generate new content (for example, creating a PowerPoint presentation from a Word document), Copilot ensures that data protection policies are preserved and inherited.

If the source Word document is labeled Highly Confidential, the generated presentation will automatically inherit the same sensitivity label. This ensures that the classification, encryption, and access restrictions applied to the original content are consistently enforced in derived content. This behavior aligns with Microsoft’s compliance and data governance model, which prevents accidental data exposure when content is reused or transformed.

Option A is incorrect because Copilot does not require manual reclassification when a label is already present on the source. Option B is incorrect because labels are not dropped during content generation. Option D is incorrect because Copilot can still generate content from labeled documents, provided the user has access.

Therefore, the presentation will be generated with the Highly Confidential label applied.

ï‚· When Microsoft 365 Copilot searches for content on the web, it uses the Bing search service.

Answer: Yes

ï‚· When Microsoft 365 Copilot searches for content on the web, it uploads the entire conversation history into the search request.

Answer: No

ï‚· When Microsoft 365 Copilot searches for content on the web, the query is anonymized and stripped of any user-identifiable information.

Answer: Yes

Microsoft 365 Copilot can use web grounding when a response requires current or public information. For web search, Copilot uses the Bing search service to retrieve relevant web results, so Statement 1 is Yes . This allows Copilot to supplement its response with up-to-date internet content when web grounding is enabled or selected.

Statement 2 is No because Copilot does not send the entire conversation history to Bing as a web search request. Instead, Copilot generates a search query based on the user’s prompt and relevant context. Sending the complete conversation would create unnecessary privacy and data exposure risk, which would conflict with Microsoft’s enterprise data protection model.

Statement 3 is Yes because Microsoft 365 Copilot is designed to protect user and tenant information during web grounding. Search queries are handled in a privacy-preserving way, with user-identifiable and tenant-identifiable information removed before the query is sent to Bing.

When building a custom agent in Microsoft 365 Copilot for marketing use cases, the required capabilities must align with the intended outputs. Marketing collateral includes written copy as well as visual assets such as logos and artwork. According to Microsoft AI Business Professional documentation, generative AI solutions that produce visual creative assets require integrated image generation capabilities powered by multimodal AI models.

Option C is correct because an image generator enables the agent to create visual marketing materials such as logos, artwork, and design elements. While templates and suggested prompts can improve usability and consistency, they do not provide the underlying capability to generate images. A code interpreter is designed for data analysis, calculations, or technical scripting tasks and is not relevant to creative marketing asset production.

Therefore, to fulfill the requirement of producing both textual and visual marketing collateral, the most essential addition to the custom agent is an image generator.

In Microsoft 365 Copilot, a notebook is a workspace intended to organize and ground related Copilot work. Microsoft guidance describes notebooks as a way to bring together multiple conversations and keep them aligned to the same set of reference materials—such as documents, notes, links, or other resources—so you don’t have to repeatedly attach or restate the same context. This improves consistency and prompt-grounding across a set of related tasks (for example, managing a project, developing a proposal, or maintaining a recurring report).

Option B correctly captures this purpose: a notebook provides a private, organized location where reference materials can be curated and reused across related Copilot conversations.

Option A is incorrect because notebooks are not primarily designed to generate email-ready transcripts of conversations. Option C is incorrect because generating a summary of interactions is a conversation-level function; notebooks are broader organizational containers for materials and related workstreams, not a “summary generator” of a single chat thread.

Therefore, the purpose of a notebook is to organize reference materials across related conversations.

Copilot in Word assists with content generation, summarization, rewriting, and structural organization. It can analyze document headings and insert a table of contents based on structured heading styles, supporting document navigation. It can also generate summaries of key points, which aligns directly with generative AI text analysis capabilities.

Custom watermark formatting and precise page margin configuration are traditional Word formatting functions and are not core generative AI capabilities of Copilot.

Therefore, the correct answers are A and C.

==================================================

a specific date range of activity

The Microsoft 365 My Account portal provides users with control over their Copilot activity history in alignment with enterprise privacy and compliance standards. When selecting Delete history , users can remove Copilot activity based on a defined time range rather than deleting only a single conversation or all activity universally.

This functionality reflects Microsoft’s commitment to transparency, user control, and responsible AI governance. Allowing deletion by date range enables organizations and individuals to manage data retention policies efficiently while maintaining compliance with regulatory frameworks such as GDPR and internal data governance policies.

The other options are incorrect because deleting a specific conversation or all conversations with a specific agent is not the primary method offered in the My Account activity deletion setting. Instead, deletion is structured around activity time periods.

This capability reinforces generative AI best practices: secure data management, lifecycle control of AI interactions, and user-directed privacy management within enterprise environments.

The correct answer is instructions because an agent’s persona—such as being warm, friendly, formal, or analytical—is defined through its behavioral and communication guidance. In Microsoft 365 Copilot, the Instructions field controls how the agent responds, including tone, style, personality, response structure, and interaction patterns.

The description typically explains what the agent does, not how it communicates. Capabilities define what tasks the agent can perform, such as summarizing documents or analyzing data. Suggested prompts provide example user queries but do not control the agent’s personality or communication style.

To establish a warm and friendly persona, the instructions should explicitly state guidance such as: “Respond in a supportive, conversational tone. Use clear and positive language. Be encouraging and approachable.” This ensures consistent tone across all interactions.

Defining persona through instructions reflects prompt engineering best practices in generative AI, enabling predictable behavior, improved user experience, and alignment with organizational communication standards within Microsoft 365 Copilot environments.

Microsoft 365 Copilot operates within the security, compliance, and identity boundaries of a Microsoft 365 tenant. Shared prompts, prompt links, and Copilot artifacts are governed by organizational access controls and tenant isolation. If a prompt is created and shared from outside your organization, cross-tenant access may not be supported depending on the sharing configuration and administrative policies.

When a user attempts to open a prompt that resides in another organization’s tenant without proper cross-tenant sharing permissions, Copilot cannot locate or validate the resource within the user’s own environment. As a result, the system displays a “Prompt not found” message.

Option B would typically result in an access or permissions error rather than the prompt being unavailable entirely. Sensitivity labels and scheduled prompts do not inherently cause a “not found” error. Therefore, the most likely cause is that the prompt exists outside your organization’s tenant boundary and is not accessible to you.

==================================================

Prompt scheduling in Microsoft 365 Copilot enables users to automate recurring AI-driven tasks such as generating reports, summaries, or updates. However, there are defined operational limits to ensure system performance, responsible AI usage, and governance compliance.

First, the most frequent interval at which a prompt can be scheduled is once per day . Scheduling more frequently, such as hourly, is not supported. This limitation prevents excessive automated execution that could strain system resources or generate unnecessary redundant outputs.

Second, a scheduled prompt has a maximum execution limit of 15 times . After reaching this threshold, the schedule must be recreated if further automation is required. This control ensures users periodically review and validate automated prompts for relevance, accuracy, and business alignment.

These constraints reflect Microsoft’s enterprise AI governance strategy: balancing productivity automation with performance optimization, cost control, compliance oversight, and responsible AI lifecycle management within Microsoft 365 Copilot environments.