practice · AI-901
AI-901 Practice Exam 3: Microsoft Azure AI Fundamentals
AI-901 practice exam 3 for Microsoft Azure AI Fundamentals: 45 original questions with explanations. Pause when needed, choose your answer, then review the reasoning.
Video summary
AI-901 practice exam 3 for Microsoft Azure AI Fundamentals: 45 original questions with explanations. Pause when needed, choose your answer, then review the reasoning.
Study notes
- Answer 45 original questions before reviewing the explanations.
- Review missed questions across responsible AI, Foundry, agents, speech, vision, and Content Understanding.
Chapters
- Introduction and how to use this practice exam
- Q1: Fairness in AI systems
- Q2: Handling uncertain AI recommendations
- Q3: Minimizing retained personal data
- Q4: Accessible alternatives to speech output
- Q5: Disclosing AI use and limitations
- Q6: Accountability for AI services
- Q7: How generative AI produces responses
- Q8: Choosing models by input and output modality
- Q9: Deployment geography and billing requirements
- Q10: Generative AI for original text
- Q11: Text analysis: entities and key concepts
- Q12: Transcribing recorded voice messages
- Q13: Generating images from written briefs
- Q14: Content Understanding field requirements
- Q15: Tokens versus words
- Q16: Model capabilities for application functions
- Q17: Selecting the intended model deployment
- Q18: Analyzing audio and visual content in video
- Q19: Analyzing opinions versus summarizing text
- Q20: Separating system instructions and user requests
- Q21: Testing prompt changes in Foundry
- Q22: Connecting with AIProjectClient in Python
- Q23: Writing prompt-agent instructions
- Q24: Selecting a prompt agent from Python
- Q25: Passing file contents to sentiment analysis
- Q26: Audio input through Chat Completions
- Q27: Speech SDK microphone recognition
- Q28: Building a vision-enabled chat request
- Q29: Text-to-image generation requests
- Q30: Image descriptions and text recognition
- Q31: Reading structured invoice analysis results
- Q32: Defining Content Understanding outputs
- Q33: Analyzing call topics and sentiment
- Q34: Handling asynchronous analysis results
- Q35: Specifying response format in prompts
- Q36: Locating a Foundry project endpoint
- Q37: Reading text from a Responses API result
- Q38: Testing prompt-agent behavior
- Q39: Maintaining context across agent requests
- Q40: Sentiment labels and confidence scores
- Q41: Requesting text and audio responses
- Q42: Configuring Speech SDK resource connections
- Q43: Follow-up questions about an image
- Q44: Extracting fields from varying document layouts
- Q45: Submitting a document URL for analysis
- Reviewing missed questions and next steps