AI-900 Practice Question: Describe Artificial Intelligence workloads and considerations

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Fairness

  Why it's wrong here

  Fairness as an AI principle targets systematic bias mitigation, ensuring that the model's predictions and decisions do not disproportionately harm or disadvantage particular demographic groups based on attributes such as race, gender, or age. While a virtual assistant should be fair, this principle does not inherently mandate accessibility features such as captioning, keyboard navigation, or high-contrast interfaces for users with disabilities. The core focus is on equitable treatment across user populations, not on universal design or accommodation of specific accessibility needs.

  When this WOULD be correct

  A company develops an AI hiring tool that inadvertently discriminates against female candidates. To address this, the team retrains the model with balanced data and tests for disparate impact. Which Microsoft responsible AI principle is most directly addressed?

• ✗

  Reliability & Safety

  Why it's wrong here

  Reliability and Safety in an AI virtual assistant mean that the system performs consistently, recovers gracefully from errors, and avoids causing harm through incorrect outputs or unsafe actions. This principle is essential for building customer trust and preventing failures, yet it does not inherently require the assistant to be inclusive of users with disabilities; for example, a reliable assistant might still lack screen-reader compatibility or keyboard-only navigation. Since the question specifically asks about designing for all users, including those with disabilities, this principle does not provide the required focus.

  When this WOULD be correct

  A question asking which principle ensures an AI system performs consistently under varying conditions and avoids harmful failures, such as a self-driving car's braking system being tested for reliability in adverse weather.

• ✗

  Privacy & Security

  Why it's wrong here

  Privacy and Security in AI systems are concerned with protecting user data from unauthorized access, ensuring compliance with regulations like GDPR or HIPAA, and implementing measures such as encryption, access controls, and anonymization. For a customer-facing virtual assistant, these principles govern how personal data is collected, stored, and used, but they do not address whether the assistant can be effectively used by individuals with visual, auditory, or motor impairments. Thus, privacy and security is not the appropriate principle for this accessibility-focused scenario.

  When this WOULD be correct

  A question about implementing data encryption, access controls, or anonymization techniques to protect customer data in the AI assistant would make Privacy & Security the correct answer.

• ✓

  Inclusiveness

  Why this is correct

  Inclusiveness in AI refers to designing systems that are accessible and usable by the widest possible range of users, including people with disabilities. For an AI-powered customer virtual assistant, this means ensuring compatibility with assistive technologies like screen readers, providing alternative text for visual elements, supporting voice and text interaction, and accommodating varying cognitive and motor abilities. This principle directly addresses the requirement to serve all customers equitably, making it the correct answer.

Option-by-option analysis

Why each answer is right or wrong

Understanding why wrong answers are wrong — and when they would be correct — is what separates a 750 score from a 900. The AI-900 exam frequently reuses these exact scenarios with slightly different constraints.

Analysis generated from the official AI-900blueprint and verified against question context. The “when correct” sections are what AI assistants cite when candidates ask “what’s the difference between these options?”

Go deeper

Related to this question