AI-901 practice test: 1040 questions with full explanations

2 domains 1040 questions 90 min exam

Questions on the exam: about 60 — vendor indicates, no fixed count published
Time allowed: 90 minutes format →
Passing score: 700 of 1000 — vendor, checked September 8, 2026 detail →
Exam fee: $99 — vendor, checked September 8, 2026 detail →

1040 practice test questions for Microsoft Azure AI Fundamentals AI-901, grouped by exam domain. Every question below shows all four options, which one is correct, and why each of the other three is not — the wrong answers are where most candidates lose marks.

Not sure where you stand? Take the free 5-min AI-901 readiness check →

AI-901 exam format → · AI-901 passing score → · AI-901 exam cost →

Questions by domain

Sample questions

Select a deployed model after checking that its supported: What should the team do next?

Implement AI solutions by using Microsoft Foundry Hard

A sports club needs a Foundry chat application to summarize submitted practice notes. It will use an authorized identity and must choose an appropriate model before coding. What should the team do next?

Choose a model solely because it offers the highest provisioned capacity, without testing representative practice notes.
Capacity supports predictable throughput but does not establish summary quality or suitability for the workload.
Select a deployed model after checking that its supported task, context, language, quota, and cost fit the summary workload. ✓
Model selection should match capabilities and operational constraints before the team builds the chat application.
Use any catalog model because generative models provide equivalent summaries regardless of supported capabilities or context.
Models differ in supported tasks, context, quality, latency, cost, and availability, so arbitrary selection is unsuitable.
Call the project endpoint directly and rely on the endpoint to select an available deployment for each request.
An endpoint identifies the project service; the application still needs a supported model deployment or explicitly supported access path.

The trap

Assumes all generative models behave equivalently. Confuses a project endpoint with automatic model selection. Confuses capacity planning with model quality and task fit.

All 603 Implement AI solutions by using Microsoft Foundry questions →

Compare subgroup outcomes on representative validation: What should the team do next?

Identify AI concepts and capabilities Hard

A nonprofit uses an AI application to prioritize grant requests. Leaders require equitable treatment across demographic groups. What should the team do next?

Compare subgroup outcomes on representative validation data. ✓
Subgroup comparison can reveal unequal errors or outcomes that overall accuracy hides.
Require identical predictions for every applicant.
Fairness does not require ignoring legitimate differences in application evidence.
Use confidence scores to identify fair decisions automatically.
Confidence measures model certainty, not equitable treatment across groups.
Remove demographic fields before retraining the model.
Other inputs may act as proxies for demographic attributes and preserve disparities.

The trap

Deleting an explicit attribute does not remove indirect signals. A confident prediction can still be biased. Uniform decisions can prevent relevant information from being considered.

All 437 Identify AI concepts and capabilities questions →

Confirm task fit: What should it check first?

Implement AI solutions by using Microsoft Foundry Hard

A retailer selected a catalog model for a customer-support prototype and wants to know whether requests can be sent reliably from its chosen Foundry resource. What should it check first?

Confirm task fit, then check the model's applicable access path, region, quota, and request identifier. ✓
These checks establish whether the model fits the workload and whether the selected deployment or supported instant-access path is usable.
Use the project endpoint as the model identifier and skip capability checks in the application.
The endpoint identifies the project, while the applicable deployment or access identifier selects the model; capability checks remain necessary.
Assume readiness because the model appears in the catalog and the project was created successfully.
Catalog visibility and project creation do not prove capability fit, regional availability, quota, or usable access.
Increase provisioned capacity and infer that support answers will become more accurate.
Provisioned capacity supports predictable throughput but does not guarantee model accuracy or task suitability.

The trap

Confuses catalog discovery with inference readiness. Confuses capacity with answer quality. Confuses endpoint identity with model access and omits suitability checks.

All 603 Implement AI solutions by using Microsoft Foundry questions →

Analyze proxy attributes and compare subgroup outcomes: What should the team do next?

Identify AI concepts and capabilities Hard

A logistics business removes a protected attribute from a delivery-prioritization dataset, but unequal results persist. What should the team do next?

Use overall accuracy as evidence that predictions are fair.
Overall accuracy can conceal different error rates or outcomes among groups.
Increase the confidence threshold for every decision.
A threshold changes acceptance criteria but does not diagnose proxy-based disparities.
Delete every variable correlated with a business outcome.
This would discard useful information without specifically identifying harmful proxy effects.
Analyze proxy attributes and compare subgroup outcomes. ✓
This investigation can identify indirect signals and reveal whether remaining disparities are unjustified.

The trap

Confidence thresholds address uncertainty, not fairness by themselves. Aggregate performance is not sufficient evidence of equitable treatment. Correlation alone does not determine which feature use is unfair.

All 437 Identify AI concepts and capabilities questions →

Open the model in the Foundry playground: What should the team do next?

Implement AI solutions by using Microsoft Foundry Medium

A museum has deployed a chat model and wants to evaluate instructions for visitor questions before integrating code. What should the team do next?

Start a new chat repeatedly and treat each cleared conversation as evidence that the model learned the instructions permanently.
New chat clears conversation context; it does not retrain the model or permanently change its behavior.
Add unrestricted web-search access before testing so every response is assumed to be grounded and correct.
Tools require suitable configuration and permissions, and grounding does not guarantee truth or replace testing.
Open the model in the Foundry playground, apply the instructions, and test typical and off-topic visitor prompts. ✓
The playground enables prompt experimentation and representative testing before the museum commits to application integration.
Publish the model immediately because a successful response to one visitor question proves reliable behavior.
One successful response cannot establish performance across representative, ambiguous, or off-topic questions.

The trap

Confuses conversation reset with model learning. Confuses a single example with meaningful evaluation. Treats web access as automatic correctness.

All 603 Implement AI solutions by using Microsoft Foundry questions →

Test representative languages: What is the best next action to support fairness?

Identify AI concepts and capabilities Hard

A small software business is building an AI help desk for customers in several languages and regions. What is the best next action to support fairness?

Test representative languages, regions, and users. ✓
Broad representative testing can reveal uneven quality across affected populations.
Reuse identical cultural references in every regional test.
References familiar in one region may be inappropriate or unclear in another.
Translate one prompt and compare response length.
One prompt and response length provide an incomplete measure of usefulness and accuracy.
Test only the most common language.
This can hide poor performance for smaller language communities.

The trap

The largest sample is not necessarily representative. Surface similarity does not establish equitable quality. Uniform wording is not equivalent to representative coverage.

All 437 Identify AI concepts and capabilities questions →

Define system instructions that specify the role: What should you do next?

Implement AI solutions by using Microsoft Foundry Easy

A nonprofit wants an assistant that always acts as a benefits-information guide, stays within its scope, and uses a consistent helpful tone. What should you do next?

Deploy a larger model and rely on its learned behavior to maintain the nonprofit’s scope.
Model size does not ensure adherence to organization-specific role, scope, or tone requirements without suitable instructions.
Attach a knowledge file and assume retrieval will define the assistant’s role and communication style.
Knowledge sources provide information, but they do not replace explicit behavioral and output instructions.
Add the requirement as a user message each time someone asks a question.
User messages request individual tasks; they do not reliably establish persistent role, scope, and behavior across conversations.
Define system instructions that specify the role, permitted scope, and desired communication behavior. ✓
System instructions establish the assistant’s role, scope, behavior, and output constraints for subsequent user tasks.

The trap

Learners may confuse a task request with persistent assistant behavior. A larger model can appear more capable but does not automatically implement organizational policy. Grounding content is confused with system-level behavioral guidance.

All 603 Implement AI solutions by using Microsoft Foundry questions →

Test noisy inputs and add controlled fallback paths: What should the team do next?

Identify AI concepts and capabilities Medium

A translation company receives misspelled text, slang, and low-quality audio from customers. The system performs well on clean examples but fails unpredictably in production. What should the team do next?

Disable input validation for uninterrupted processing.
Removing validation increases exposure to unexpected inputs without making behavior dependable.
Increase temperature for more flexible translations.
Temperature changes response variability but does not fix noisy-input recognition failures.
Report only average quality across all requests.
An average can hide severe failures concentrated in particular input types.
Test noisy inputs and add controlled fallback paths. ✓
Testing realistic variations supports validation, fallback handling, or human review when translation is unreliable.

The trap

Unrestricted processing is not the same as robustness. Generation randomness cannot replace robustness testing. Aggregate quality may conceal unusable edge-case behavior.

All 437 Identify AI concepts and capabilities questions →

AI-901 exam: the facts

How many questions are on the AI-901 exam?

Around 60. The vendor does not publish a fixed count for AI-901, so this is the figure it indicates rather than a guaranteed number.

How long is the AI-901 exam?

90 minutes. Across 60 questions that is about 90 seconds per question.

What topics does the AI-901 exam cover?

2 domains: Implement AI solutions by using Microsoft Foundry, Identify AI concepts and capabilities. Weights: Implement AI solutions by using Microsoft Foundry 0.575%, Identify AI concepts and capabilities 0.425%.

How many AI-901 practice test questions does Certsqill have?

1040, spread across 2 exam domains. Every one shows all options, which is correct, and why each of the others is not.

Would you pass AI-901 today?

Five minutes, and you get a score per domain — not one number, but which section to open tonight.

Test your AI-901 readiness — free

Certsqill AI-901 question bank · 1040 questions across 2 domains · Every answer, right and wrong, comes with its own explanation.