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.

• ✗

  AI systems with publicly available source code that anyone can inspect

  Why it's wrong here

  Open-source AI is about code transparency—anyone can inspect the training code or model architecture—but it does not enable a system to justify its individual predictions in human-understandable terms. A deep learning model with public source code is still a black box because its internal weights and transformations are not inherently interpretable. Explainable AI specifically addresses the need to communicate decision rationale to affected parties, not to publish code for public audit.

• ✓

  AI systems that can explain their decisions in understandable terms — required for regulatory compliance

  Why this is correct

  This is the core definition of explainable AI: producing decision-specific justifications in plain, human-comprehensible language that affected stakeholders can understand and challenge. It is a regulatory necessity under frameworks like the EU's GDPR (Article 22) and the EU AI Act, which grant individuals a right to meaningful explanations for consequential automated decisions such as credit scoring, hiring, or medical triage. These explanations must cover the input factors and reasoning that led to the outcome, enabling accountability and contestability, rather than merely publishing code or data.

• ✗

  AI models that are simple enough for non-experts to rebuild from scratch

  Why it's wrong here

  Model simplicity (e.g., linear regression, small decision trees) is one route to intrinsically interpretable models, but explainable AI is the broader field that also provides explanation techniques like SHAP, LIME, and counterfactuals for complex black-box systems. Being able to rebuild a model from scratch may indicate transparency, but it does not guarantee that the model's individual decisions can be understood by non-experts. XAI's goal is to make the logic behind specific outcomes comprehensible to stakeholders, not just to make the model architecture reproducible.

• ✗

  AI systems that automatically explain errors in user-submitted code

  Why it's wrong here

  This describes developer tooling such as AI-assisted debuggers or static analyzers that generate natural-language notes about bugs in user-written code, not the AI system's own reasoning. Explainable AI is concerned with the system explaining its decisions to the individuals or regulators affected by those decisions, such as why a loan application was denied. The direction of explanation differs: XAI explains the AI's internal decision process, not external code defects.

Go deeper

Related to this question