View Full ISTQB CT-GenAI Exam Dumps and Practice Test Dumps.

 

Question 361

A tester asks GenAI to generate test cases for a new feature, but the prompt does not identify the feature’s supported user roles. What is the main concern?

1. The generated tests may miss role-specific behavior and access conditions
2. The model will automatically determine all user roles
3. The generated tests will always be performance tests
4. The model will refuse to generate any tests

Correct Answer: 1

Explanation

User roles can significantly affect application behavior and access rights. If role information is missing, GenAI may generate only generic scenarios and overlook differences between administrators, standard users, guests, or other roles. This can result in incomplete functional and authorization coverage. The tester should provide relevant role definitions and expected permissions when they are necessary for the testing objective. Generated scenarios should then be checked against the approved access-control and functional requirements. GenAI can help expand role-based scenarios, but it should not invent permissions or assume that all users have identical capabilities.

Question 362

Which approach can help a tester obtain more useful results when a GenAI response is too general?

1. Remove all context from the prompt
2. Add relevant constraints, examples, and specific expected output requirements
3. Ask the model to produce a longer response without changing the context
4. Repeat exactly the same prompt indefinitely

Correct Answer: 2

Explanation

A response that is too general may indicate that the prompt does not provide enough task-specific direction. Adding relevant constraints, examples, project terminology, expected output fields, and clear objectives can help GenAI focus on the required testing activity. Simply requesting more words does not necessarily improve relevance or accuracy. Removing context can make the problem worse. Prompt refinement should be iterative, with testers evaluating whether each change produces more useful and reliable results. Even after refinement, the output must be validated against authoritative requirements because better prompting improves usefulness but does not guarantee correctness.

Question 363

A tester asks GenAI to identify possible regression tests after a change to a payment calculation module. Which information is most useful?

1. The tester’s preferred documentation style
2. The number of developers working on the module
3. Changed functionality, dependencies, affected business rules, and existing test coverage
4. The application’s logo and branding guidelines

Correct Answer: 3

Explanation

Regression-test selection depends on understanding what changed and what other behavior may be affected. Information about modified functionality, dependencies, business rules, interfaces, and existing coverage helps GenAI identify potentially impacted tests. The generated suggestions should be compared with technical impact analysis and the current test suite because the model may overlook hidden dependencies or suggest unrelated tests. Regression testing should focus on meaningful risk rather than simply generating a large number of cases. GenAI can accelerate identification of candidates, but testers remain responsible for determining the final regression scope.

Question 364

A GenAI tool produces different test cases each time the same prompt is submitted. What characteristic of generative systems does this illustrate?

1. Guaranteed determinism
2. Automatic requirement traceability
3. Probabilistic or non-deterministic output generation
4. Automatic defect confirmation

Correct Answer: 3

Explanation

Generative AI systems can produce different outputs from the same or very similar prompts because generation is often probabilistic. This characteristic means that testers should not assume that repeating a prompt will always produce an identical set of test cases. For important activities, teams may need appropriate review, recording, and evaluation procedures. Variability can sometimes provide useful alternative scenarios, but it can also introduce inconsistencies. Testers should therefore assess generated outputs against defined requirements and quality criteria rather than assuming that one generated response represents a complete or authoritative answer.

Question 365

A tester asks GenAI to generate tests for a service that depends on an external payment provider. What should be included in the testing context?

1. Only the service name
2. The external dependency behavior, failure conditions, and integration requirements
3. The tester’s preferred programming language
4. The number of pages in the requirements document

Correct Answer: 2

Explanation

External dependencies can introduce important success and failure conditions that should be represented in testing. The prompt should provide relevant integration requirements, expected responses, dependency behavior, timeout conditions, unavailable-service scenarios, and any applicable error-handling rules. Without this information, GenAI may generate only successful interaction scenarios and overlook important integration failures. Generated tests should be checked against the actual service contract and approved requirements. Testers should also consider whether external-service behavior can be simulated or stubbed during testing. Good context helps GenAI produce more comprehensive integration scenarios while maintaining appropriate human validation.

Question 366

A tester receives a GenAI-generated test containing a fabricated requirement identifier. What is the appropriate action?

1. Verify the identifier against the project’s requirements repository
2. Create a new requirement automatically
3. Use the identifier because it looks realistic
4. Remove all requirement traceability

Correct Answer: 1

Explanation

A realistic-looking requirement identifier does not prove that the requirement exists. The tester should verify the identifier against the authoritative requirements repository or approved documentation. If it cannot be found, the generated traceability information should not be treated as valid. GenAI can fabricate identifiers when attempting to create structured outputs, especially when examples or incomplete context are provided. Verification prevents false traceability links from entering the test-management system. Where traceability is required, each test should ultimately be connected to a real and approved requirement, risk, or other authoritative source.

Question 367

A testing team wants to use GenAI for exploratory testing. Which use is most appropriate?

1. Generate hypotheses and unusual scenarios for testers to investigate
2. Automatically declare every generated scenario as a defect
3. Replace all exploratory testers with generated scripts
4. Ignore observations made during execution

Correct Answer: 1

Explanation

GenAI can support exploratory testing by suggesting alternative user behaviors, unusual combinations, edge conditions, and investigation questions. These suggestions can provide testers with ideas that may broaden exploration. However, generated scenarios are not automatically defects and should be investigated using the actual application and available evidence. Exploratory testing relies heavily on tester observation, learning, and adaptation, so human participation remains important. GenAI can act as an idea-generation assistant rather than an autonomous decision-maker. Testers should document meaningful findings and use their judgment to determine whether observed behavior represents a defect or another relevant result.

Question 368

A tester wants generated test cases to contain only fields required by the team’s test-management system. Which prompt technique is most helpful?

1. Ask for unlimited free-form output
2. Specify an exact output schema and required fields
3. Remove all examples
4. Ask the model to write a general essay about testing

Correct Answer: 2

Explanation

An explicit output schema can help constrain generated responses to the structure required by a test-management system. The prompt can identify mandatory fields, field names, permitted formats, and any rules for missing information. Examples can also demonstrate the expected structure when appropriate. However, generated output should still be validated before import because the model may omit fields, produce invalid values, or introduce unsupported information. Structured prompting is particularly useful when GenAI output will be processed by another system. It improves consistency and reduces manual formatting effort without eliminating the need for quality and content validation.

Question 369

A tester asks GenAI to generate tests from a business process containing several decision points. Which technique can help ensure important paths are considered?

1. Ask for scenarios covering meaningful combinations of the decision outcomes
2. Test only the first decision point
3. Remove all alternative paths from the prompt
4. Generate one test regardless of the number of decisions

Correct Answer: 1

Explanation

Business processes with multiple decision points can create numerous possible paths. Asking GenAI to consider meaningful combinations of decision outcomes can help identify important positive, negative, and exception paths. The tester should provide the actual business rules and constraints so that generated combinations remain grounded in the process. Not every mathematically possible combination necessarily requires testing, so risk and coverage objectives should guide selection. Generated paths should also be reviewed for duplicates, infeasible combinations, and missing conditions. GenAI can help explore the possibility space, but testers must determine which paths provide appropriate and valuable coverage.

Question 370

A tester uses GenAI to suggest test automation libraries. What should be done before adding a suggested library to the project?

1. Install it immediately because the model recommended it
2. Verify its suitability, security, licensing, compatibility, and organizational approval
3. Assume every suggested library is officially supported
4. Ignore all project dependency policies

Correct Answer: 2

Explanation

GenAI may suggest libraries that are outdated, unsuitable, insecure, incompatible, or subject to licensing conditions that do not fit the project. Before adoption, the tester or development team should evaluate the library against technical requirements, security policies, licensing rules, supported versions, maintenance status, and organizational approval processes. The model’s recommendation should be treated as a candidate rather than an authoritative decision. This is particularly important because generated recommendations may contain fabricated or outdated information. Proper dependency review protects the project from introducing unnecessary technical, security, and compliance risks.

Question 371

A tester wants to compare two sets of GenAI-generated test cases. Which comparison is most meaningful?

1. Compare their coverage, correctness, redundancy, and relevance to the requirements
2. Compare only the number of words
3. Choose the set generated first
4. Choose the set with the longest test descriptions

Correct Answer: 1

Explanation

The value of a generated test set should be assessed using meaningful testing criteria. Coverage shows whether relevant requirements and risks are addressed, correctness shows whether expected behavior is accurate, redundancy identifies unnecessary duplication, and relevance determines whether scenarios apply to the system under test. Word count and generation order do not provide meaningful evidence of test quality. A structured comparison can help teams determine which generated artifacts require fewer corrections and which provide stronger testing value. Human review remains necessary because some important quality characteristics depend on project-specific knowledge that may not be available to GenAI.

Question 372

A tester asks GenAI to create negative tests for a file-upload feature. Which scenario is appropriate?

1. Uploading only the documented valid file type
2. Uploading a prohibited or unsupported file type
3. Opening the application’s help page
4. Changing the application’s language

Correct Answer: 2

Explanation

Negative testing checks how the system behaves when invalid, unexpected, or prohibited inputs are provided. For a file-upload feature, unsupported file types are a relevant negative condition. Other useful cases may include excessive file size, malformed content, invalid filenames, duplicate uploads, or unauthorized access, depending on the requirements. GenAI can suggest such conditions when the prompt includes the file-upload rules and security expectations. Each generated case should be checked against the actual requirements so that testers do not create unsupported assumptions. Negative testing helps determine whether the application handles invalid conditions safely and predictably.

Question 373

A tester wants GenAI to generate tests for a regulated system. Which factor should receive particular attention?

1. Traceability, review records, approved sources, and applicable governance requirements
2. Generating the largest possible number of tests
3. Avoiding documentation to save time
4. Allowing unreviewed model output into the final test suite

Correct Answer: 1

Explanation

Regulated systems may require stronger controls around traceability, evidence, approvals, documentation, and accountability. When GenAI is used, teams should understand how generated artifacts were produced, what authoritative sources were used, who reviewed them, and what changes were made. Applicable organizational and regulatory requirements should guide the process. Quantity alone does not establish compliance or testing quality. Unreviewed generated content can create risks if it contains unsupported assumptions or inaccurate information. GenAI may improve productivity in regulated testing, but governance and human oversight remain important parts of the overall testing process.

Question 374

A tester notices that GenAI repeatedly generates scenarios using the same customer type. What could help improve diversity of generated scenarios?

1. Explicitly request relevant customer categories and varied representative conditions
2. Ask the model to repeat the same scenario
3. Remove customer context entirely
4. Restrict the output to one example

Correct Answer: 1

Explanation

If generated scenarios repeatedly focus on one customer type, the prompt can explicitly identify relevant categories and request representative variation across them. The tester should provide legitimate business distinctions such as customer segments, account states, permissions, or applicable service plans. This can help broaden scenario diversity while remaining grounded in the application’s requirements. The goal is not to generate arbitrary variety but to cover meaningful differences that could affect behavior. Generated scenarios should still be reviewed for relevance and feasibility. Explicit diversity requirements can reduce repetitive outputs and help testers explore conditions that might otherwise receive insufficient attention.

Question 375

A tester asks GenAI to convert a user story into acceptance tests. What should be checked first?

1. Whether the user story contains clear and testable acceptance criteria
2. Whether the model uses advanced vocabulary
3. Whether the generated tests contain many steps
4. Whether the prompt is longer than the user story

Correct Answer: 1

Explanation

Acceptance tests depend on having clear expectations for what constitutes successful behavior. If a user story contains ambiguous or incomplete acceptance criteria, GenAI may fill the gaps with assumptions. The tester should first assess whether the story defines observable, testable conditions and clarify missing information where necessary. Once the criteria are sufficiently clear, GenAI can help transform them into structured acceptance tests. Generated tests should then be reviewed for traceability and correctness. Prompt length or sophisticated vocabulary does not compensate for unclear requirements. Clear acceptance criteria provide a stronger foundation for both manual and automated test generation.

Question 376

A tester receives a GenAI-generated test that includes credentials in the example data. What should the tester do?

1. Verify that the credentials are safe, synthetic, and appropriate for testing
2. Use real production credentials to make the test realistic
3. Publish the credentials in the test report
4. Store the credentials directly in source code

Correct Answer: 1

Explanation

Credentials require careful handling because exposing real authentication information can create serious security risks. Testers should determine whether generated credentials are synthetic and whether they are appropriate for the intended test environment. Real production credentials should not be introduced merely to make examples realistic. Where test accounts are required, they should follow the organization’s credential-management practices. Generated automation should also avoid hard-coded secrets when secure alternatives are available. Reviewing generated test data for credentials and other sensitive information is an important part of secure GenAI-assisted testing.

Question 377

A tester wants to use GenAI to rewrite technical test results for business stakeholders. What should the prompt specify?

1. The target audience, required level of detail, and important information to preserve
2. Only the model’s preferred writing style
3. That all technical facts should be removed
4. That the model should invent business impacts

Correct Answer: 1

Explanation

Different audiences require different levels of detail and terminology. A prompt intended for business stakeholders should identify the audience, desired communication style, level of technical detail, and facts that must be preserved. GenAI can then transform technical information into clearer business-oriented language without unnecessarily changing the underlying meaning. Testers should verify the rewritten content to ensure that important facts, limitations, and results have not been distorted. The model should not invent business impacts that are not supported by evidence. Audience-aware prompting is therefore useful for communication while factual accuracy remains the responsibility of the reviewer.

Question 378

A tester asks GenAI to identify security risks in generated automation code. What limitation should be considered?

1. GenAI may miss vulnerabilities or incorrectly classify code as safe
2. GenAI always detects every security vulnerability
3. Generated security analysis is equivalent to penetration testing
4. Human security review is never required

Correct Answer: 1

Explanation

GenAI can assist with identifying potentially insecure patterns, but it cannot guarantee complete security analysis. It may overlook vulnerabilities, misunderstand application context, or incorrectly classify a code pattern as safe or unsafe. Security review should therefore use appropriate additional techniques, such as static analysis, dependency scanning, code review, and authorized security testing where applicable. Generated suggestions can help reviewers focus their attention but should not be treated as comprehensive assurance. This limitation is particularly important for security-sensitive automation that interacts with credentials, external systems, privileged accounts, or production-like environments.

Question 379

A testing team wants to ensure that GenAI-generated tests remain aligned with changing requirements. What practice is most useful?

1. Periodically review generated artifacts against the latest approved requirements
2. Keep generated tests unchanged forever
3. Use only the original prompt regardless of requirement changes
4. Remove requirement traceability after generation

Correct Answer: 1

Explanation

Requirements can change over the life of a project, making previously generated tests potentially outdated. Periodic review against the latest approved requirements helps identify tests that need modification, removal, or additional coverage. Traceability makes this review easier because testers can identify which artifacts are affected by a requirement change. Reusing an old prompt without updating its context may produce obsolete results. GenAI can assist with identifying potentially affected tests, but the current approved requirements remain authoritative. Continuous alignment is especially important for regression suites and other test assets maintained over multiple releases.

Question 380

Which statement best describes the role of human oversight when GenAI is used in software testing?

1. Humans remain responsible for reviewing and validating important generated outputs
2. Humans are needed only when the model produces an error message
3. Human review is unnecessary when the prompt is detailed
4. GenAI automatically assumes responsibility for testing decisions

Correct Answer: 1

Explanation

Human oversight remains important because GenAI can produce incorrect, incomplete, outdated, biased, or unsupported information. Testers and other responsible professionals should review generated requirements interpretations, test cases, automation code, defect reports, summaries, and other artifacts according to their risk and purpose. A detailed prompt can improve output quality but cannot eliminate the possibility of errors. Human reviewers provide domain knowledge, verify evidence, resolve ambiguity, and make appropriate testing decisions. GenAI is therefore best used as an assistant that supports testing activities while accountability for important decisions and final artifacts remains with the responsible people.