Frequently Asked Questions
What this playlist answers
9 questions this playlist covers — skim
them for a quick takeaway, or dive into the videos above for the full picture.
This six-part learning path teaches you how to use generative AI in real workplace tasks. It covers core AI concepts, prompt engineering, choosing between AI augmentation and automation, using Microsoft Copilot, and applying AI to business use cases such as data analysis, marketing, sales, and customer support, along with responsible AI practices.
Generative AI is a type of artificial intelligence that creates new content such as text, images, audio, video, and code based on patterns learned from data. Traditional AI usually focuses on analyzing data, making predictions, or completing predefined tasks, while generative AI focuses on producing new outputs.
Generative AI can support research, document summarization, brainstorming, content creation, data analysis, coding, meeting summaries, and workflow organization. It can either augment your work by supporting human judgment or automate lower-risk, repetitive tasks.
Prompt engineering is the practice of writing clear, structured instructions that guide generative AI toward a desired output. Effective prompts start with a clear goal, add relevant context, explain the task, and specify the desired format, and can include examples through zero-shot, one-shot, or few-shot prompting.
AI augmentation uses generative AI to support your work while keeping human judgment and decision-making in the loop, which suits complex or higher-stakes tasks. AI automation uses AI to complete repetitive or lower-risk tasks with less human involvement. The right approach depends on the task's complexity, risk, and need for human judgment.
Microsoft Copilot is an AI assistant built into Microsoft tools like Teams that helps users summarize meetings, identify action items, analyze data, create content, and manage tasks. It brings generative AI directly into everyday workflows to improve productivity and collaboration.
Generative AI can analyze datasets to identify patterns, segments, and sentiment, prioritize high-value sales prospects and personalize outreach, and help draft tailored customer support responses. Outputs should be treated as a starting point and reviewed by people before use.
AI hallucinations occur when a generative AI system produces information that sounds convincing but is inaccurate, because the model predicts likely outputs rather than verifying truth. You can reduce the risk by matching the level of human verification to the stakes of the task and independently checking important outputs.
AI outputs depend heavily on data quality, so strong data governance helps reduce bias, protect sensitive information, and improve accuracy. Users should follow organizational privacy policies, limit sensitive data shared with AI tools, and keep human review as part of AI-assisted workflows.