ISC CCSP Practice Test 25
Question 240
Before changing production, which focused diagnostic is most useful for identifying the true dependency? A cloud coding team has never practiced identifying server-side request forgery and deploys an image-fetching API that accepts arbitrary URLs. Select the ONE best answer.
- Inspect retrieval payloads, tool authorization, model prompts, adversarial test cases and audit logs of attempted actions.
- Test URL parser edge cases, redirect chains, DNS resolution and blocked metadata-network attempts.
- Review dependency ownership, license terms, release signatures, active maintainers and runtime access attempts.
- Inspect CASB connector scopes, data categories, employee notice requirements, access logs and privacy impact assessment.
Explanation
Correct Answer
B:SSRF can cause server-side components to access privileged network locations that external callers cannot directly reach. Add cloud-specific secure coding training and implement strict destination validation with metadata-service and internal-network defenses.
Incorrect Answers
A:Untrusted model context and external documents can influence AI output but must not be treated as authorization to perform actions. The focused check for the reported condition is: Test URL parser edge cases, redirect chains, DNS resolution and blocked metadata-network attempts.
C:Apply third-party software intake controls including vendor and license review, source provenance, SCA and sandboxed behavior testing. The focused check for the reported condition is: Test URL parser edge cases, redirect chains, DNS resolution and blocked metadata-network attempts.
D:Cloud access brokerage provides visibility and controls but must still satisfy privacy, proportionality and authorization requirements. The focused check for the reported condition is: Test URL parser edge cases, redirect chains, DNS resolution and blocked metadata-network attempts.
Question 241
The change record requires diagnosis, intervention, and verification. Which full plan is most defensible? A data science group deploys an AI model through a provider API but assumes provider infrastructure controls also validate customer prompts and sensitive training uploads. Select the ONE best answer.
- Investigate: Diagram DNS, identity, cross-cloud routing, KMS, management APIs, and the outcome of a single-provider isolation drill. Remediate: Map hybrid and multi-cloud control planes, trust relationships, routing, and outage dependencies before declaring independent resilience. Verify: Failure testing of one provider maintains required workflows using independent authorized identity and connectivity dependencies.
- Investigate: Review model purpose, affected users, data categories, fairness test records, decision logs, and legal obligations. Remediate: Institute an accountable AI risk process covering data minimization, human review, impact testing and jurisdiction-appropriate legal assessment. Verify: Representative outcome tests and appeal workflows show reviewable decisions and documented controls for sensitive personal information.
- Investigate: Review model service terms, inference permissions, training-retention settings, prompt logs, and sensitive dataset lineage. Remediate: Document the AI-as-a-service responsibility boundary and implement customer data validation, prompt handling, logging, and access governance. Verify: Red-team test requests cannot expose other tenant data and uploaded datasets meet approved privacy and validation checks.
- Investigate: Identify provider-specific APIs, identity mappings, credential models, alternate implementations, and realistic migration effort. Remediate: Define abstraction, federation and exit patterns that balance necessary native services against documented portability requirements. Verify: A pilot workload authenticates on its alternate environment with a documented migration process and acceptable control equivalence.
Explanation
Correct Answer
C:Managed AI endpoints may secure model infrastructure while customers still control their submitted data, app behavior, and access policies. Red-team test requests cannot expose other tenant data and uploaded datasets meet approved privacy and validation checks.
Incorrect Answers
A:Diagram DNS, identity, cross-cloud routing, KMS, management APIs, and the outcome of a single-provider isolation drill. Remediate: Map hybrid and multi-cloud control planes, trust relationships, routing, and outage dependencies before declaring independent resilience. Verify: Failure testing of one provider maintains required workflows using independent authorized identity and connectivity dependencies. These steps address a different operational problem. They do not connect diagnosis, correction and acceptance criteria for the current requirement. The required change is: Document the AI-as-a-service responsibility boundary and implement customer data validation, prompt handling, logging, and access governance.
B:Review model purpose, affected users, data categories, fairness test records, decision logs, and legal obligations. Remediate: Institute an accountable AI risk process covering data minimization, human review, impact testing and jurisdiction-appropriate legal assessment. Verify: Representative outcome tests and appeal workflows show reviewable decisions and documented controls for sensitive personal information. These steps address a different operational problem. They do not connect diagnosis, correction and acceptance criteria for the current requirement. The required change is: Document the AI-as-a-service responsibility boundary and implement customer data validation, prompt handling, logging, and access governance.
D:Identify provider-specific APIs, identity mappings, credential models, alternate implementations, and realistic migration effort. Remediate: Define abstraction, federation and exit patterns that balance necessary native services against documented portability requirements. Verify: A pilot workload authenticates on its alternate environment with a documented migration process and acceptable control equivalence. These steps address a different operational problem. They do not connect diagnosis, correction and acceptance criteria for the current requirement. The required change is: Document the AI-as-a-service responsibility boundary and implement customer data validation, prompt handling, logging, and access governance.
Question 242
Which evidence-gathering check most directly narrows the cause of this incident? A SaaS audit team marks all provider security controls complete after receiving one report, ignoring complementary user entity controls assigned to customers. Select the ONE best answer.
- Review processing purposes, categories of personal information, recipients, consent or lawful basis and risk mitigations.
- Inspect fields, reidentification context, data-use agreements, legal definitions and access control obligations.
- Review SOC complementary controls, customer IAM policies, test results, control ownership and risk exceptions.
- Review cyber insurance coverage, MSA incident terms, notification triggers, regulator timelines and provider response simulation.
Explanation
Correct Answer
C:Even a positive provider audit opinion can depend on customer-implemented identity, configuration and data governance controls. Map provider assurance controls and customer complementary obligations to evidence, responsible owners and periodic testing.
Incorrect Answers
A:New purposes and processor access can change privacy risk even when the original customer consent and storage service remain unchanged. The focused check for the reported condition is: Review SOC complementary controls, customer IAM policies, test results, control ownership and risk exceptions.
B:Contractual confidentiality and statutory privacy categories overlap but are not identical, and identifiers can be indirect. The focused check for the reported condition is: Review SOC complementary controls, customer IAM policies, test results, control ownership and risk exceptions.
D:Insurance and service contracts serve different purposes; coverage does not ensure timely cooperation or satisfy legal notification deadlines. The focused check for the reported condition is: Review SOC complementary controls, customer IAM policies, test results, control ownership and risk exceptions.
Question 243
The operator needs proof rather than a guess. Which investigation should come first? A CASB deployment identifies unsanctioned SaaS use but has permission to read every employee's private content without a documented justification. Select the ONE best answer.
- Inspect CASB connector scopes, data categories, employee notice requirements, access logs and privacy impact assessment.
- Inspect retrieval payloads, tool authorization, model prompts, adversarial test cases and audit logs of attempted actions.
- Inspect federated login configuration, local user inventory, SCIM events, active sessions and negative deprovisioning tests.
- Review data flow diagrams, trust boundaries, service tokens, privilege levels and threat-to-control traceability.
Explanation
Correct Answer
A:Cloud access brokerage provides visibility and controls but must still satisfy privacy, proportionality and authorization requirements. Limit CASB collection to lawful approved signals, apply data minimization and enforce role-based investigator access. This check separates the suspected dependency before a risky change.
Incorrect Answers
B:Untrusted model context and external documents can influence AI output but must not be treated as authorization to perform actions. The focused check for the reported condition is: Inspect CASB connector scopes, data categories, employee notice requirements, access logs and privacy impact assessment.
C:SSO integration does not automatically revoke alternate local credentials unless identity lifecycle and fallback paths are governed. The focused check for the reported condition is: Inspect CASB connector scopes, data categories, employee notice requirements, access logs and privacy impact assessment.
D:Cloud applications face spoofing, tampering, repudiation, disclosure, denial of service and privilege escalation across internal interfaces. The focused check for the reported condition is: Inspect CASB connector scopes, data categories, employee notice requirements, access logs and privacy impact assessment.
Question 244
Which acceptance evidence would show that the original fault or risk was actually remediated? A retailer replaces primary card numbers with tokens but puts the token-vault detokenization API behind an unrestricted service account. Select the ONE best answer.
- A compromised low-privilege reporting service cannot retrieve original values and vault access is attributable in audit events.
- An administrator attempting to erase their own test event is denied and preserved evidence remains verifiable for review.
- The mapped analytics tables have a named owner, appropriate classification and tested retention enforcement.
- Held records remain protected and retrievable with an auditable hold, while unrelated records continue their approved lifecycle.
Explanation
Correct Answer
A:Tokenization reduces exposure only when the mapping service and ability to reverse tokens are separately secured. It validates the controlled change: Restrict token-vault access to approved workloads, isolate mapping data and log or approve sensitive detokenization operations.
Incorrect Answers
B:A log useful for investigation needs evidence integrity and privilege separation from subjects of the recorded activity. This is a different acceptance outcome; the required result here is: A compromised low-privilege reporting service cannot retrieve original values and vault access is attributable in audit events.
C:Derived records may remain sensitive even when their original source table is not directly exposed. This is a different acceptance outcome; the required result here is: A compromised low-privilege reporting service cannot retrieve original values and vault access is attributable in audit events.
D:The alternative observation is: Held records remain protected and retrievable with an auditable hold, while unrelated records continue their approved lifecycle. The acceptance test for this case is: A compromised low-privilege reporting service cannot retrieve original values and vault access is attributable in audit events.
Question 245
Which action directly resolves the core infrastructure requirement without relying on an unrelated workaround? A global bank signs a master service agreement that does not state whether it can retrieve customer data and encryption keys at termination. Select the ONE best answer.
- Assess applicable student-record privacy obligations and contractually govern vendor access, permitted use and disclosures.
- Negotiate clear data portability, deletion, key custody, assistance, timing and verification clauses before production lock-in.
- Map provider assurance controls and customer complementary obligations to evidence, responsible owners and periodic testing.
- Perform an appropriate privacy impact assessment, data-flow update, processor due diligence and mitigation review before activation.
Explanation
Correct Answer
B:Cloud exit viability depends on practical rights and technical mechanisms to recover data and maintain control at contract termination. Before making it, verify: Review MSA, SOW, exit schedules, storage formats, key agreements, egress fees and a tested export plan.
Incorrect Answers
A:Assess applicable student-record privacy obligations and contractually govern vendor access, permitted use and disclosures. Here the decisive dependency is: Cloud exit viability depends on practical rights and technical mechanisms to recover data and maintain control at contract termination. Use the specified intervention: Negotiate clear data portability, deletion, key custody, assistance, timing and verification clauses before production lock-in.
C:That option would map provider assurance controls and customer complementary obligations to evidence, responsible owners and periodic testing, but it addresses a different concern. Cloud exit viability depends on practical rights and technical mechanisms to recover data and maintain control at contract termination.
D:Perform an appropriate privacy impact assessment, data-flow update, processor due diligence and mitigation review before activation. Here the decisive dependency is: Cloud exit viability depends on practical rights and technical mechanisms to recover data and maintain control at contract termination. Use the specified intervention: Negotiate clear data portability, deletion, key custody, assistance, timing and verification clauses before production lock-in.
Question 246
Which observed outcome provides the strongest verification for this requirement? A legacy application concatenates user search text into SQL statements running against a managed cloud database. Select the ONE best answer.
- A compromised app identity cannot read unrelated tables and approved application queries succeed with audited access.
- The departed user fails all sign-in paths while approved break-glass accounts remain separately controlled and audited.
- A rotated secret reaches authorized running workloads without publishing it in the image or registry metadata.
- Malicious query input is treated as data, the database returns only authorized records and DAST tests cannot alter query structure.
Explanation
Correct Answer
D:SQL injection follows unsafe command construction and remains possible even when the database is a managed cloud service. It validates the controlled change: Use parameterized queries, least-privileged database identities and tests for unsafe input handling before release.
Incorrect Answers
A:A WAF inspects a subset of application traffic and cannot substitute for database authorization or internal service controls. This is a different acceptance outcome; the required result here is: Malicious query input is treated as data, the database returns only authorized records and DAST tests cannot alter query structure.
B:SSO integration does not automatically revoke alternate local credentials unless identity lifecycle and fallback paths are governed. This is a different acceptance outcome; the required result here is: Malicious query input is treated as data, the database returns only authorized records and DAST tests cannot alter query structure.
C:Container images are widely replicated artifacts and should not embed long-lived credentials or sensitive runtime configuration. This is a different acceptance outcome; the required result here is: Malicious query input is treated as data, the database returns only authorized records and DAST tests cannot alter query structure.
Question 247
Which technical principle best explains why the problem requires a targeted response? A finance group assesses only data-theft risk and ignores denial-of-service effects on a cloud payment API during peak periods. Select the ONE best answer.
- Cloud infrastructure risk includes availability and operational disruption as well as unauthorized disclosure.
- Cloud platform risk assessment depends on understanding where the provider's responsibilities end and tenant configuration begins.
- A backup being available does not guarantee the workload can be restored within the business's acceptable downtime.
- Cloud physical security remains important even when customers cannot directly inspect or operate the provider facility.
Explanation
Correct Answer
A:Evidence of correction: An approved load or failure simulation meets the business threshold and the documented risk register includes outage scenarios.
Incorrect Answers
B:The alternative governing principle is: Cloud platform risk assessment depends on understanding where the provider's responsibilities end and tenant configuration begins. The mechanism tested by the reported case is: Cloud infrastructure risk includes availability and operational disruption as well as unauthorized disclosure.
C:The alternative governing principle is: A backup being available does not guarantee the workload can be restored within the business's acceptable downtime. The mechanism tested by the reported case is: Cloud infrastructure risk includes availability and operational disruption as well as unauthorized disclosure.
D:The alternative governing principle is: Cloud physical security remains important even when customers cannot directly inspect or operate the provider facility. The mechanism tested by the reported case is: Cloud infrastructure risk includes availability and operational disruption as well as unauthorized disclosure.
Question 248
Which evidence-gathering check most directly narrows the cause of this incident? A cloud network team permits broad lateral access because guest NICs appear isolated at the hypervisor while routing connects all application segments. Select the ONE best answer.
- Trace network routes, security groups, flow logs, hypervisor port groups and service dependency inventory.
- Review routing tables, network ACLs, network capacity, NFS export rules and performance test traces.
- Examine KMS event logging options, test denials, monitoring rules, event retention and time consistency.
- Review role grants, task requirements, privilege-use logs, escalation paths and cross-project negative tests.
Explanation
Correct Answer
A:Virtual NIC boundaries do not prevent lateral movement through permitted shared networks and routed infrastructure. Map east-west flows and enforce workload-level segmentation and route controls matched to documented business dependencies.
Incorrect Answers
B:Cloud platform security requires deliberate separation and capacity planning for service-to-service and storage network dependencies. The focused check for the reported condition is: Trace network routes, security groups, flow logs, hypervisor port groups and service dependency inventory.
C:Audit mechanisms must record failures and attempted misuse, not only changes that successfully modified protected systems. The focused check for the reported condition is: Trace network routes, security groups, flow logs, hypervisor port groups and service dependency inventory.
D:Least privilege and separation of duties reduce impact from credential misuse without preventing properly authorized operations. The focused check for the reported condition is: Trace network routes, security groups, flow logs, hypervisor port groups and service dependency inventory.
Question 249
A change engineer must select a justified corrective intervention. Which should be approved? A provider offers deletion of a virtual disk while the customer assumes the cloud operator has erased every underlying flash block immediately. Select the ONE best answer.
- Perform privacy threat assessment, minimize training attributes and test model outputs for memorization or inference leakage before release.
- Classify prompt and output artifacts, redact sensitive fields and apply retention, access and regional restrictions consistent with source data.
- Plan long-term archive format, key custody, integrity checks and periodic restoration tests before decommissioning dependencies.
- Confirm the provider's sanitization and media-handling assurances and use controlled cryptographic erase where appropriate.
Explanation
Correct Answer
D:Logical deletion and physical media sanitization are distinct processes whose evidence depends on the service and control boundary. This is the targeted change for the reported environment.
Incorrect Answers
A:Perform privacy threat assessment, minimize training attributes and test model outputs for memorization or inference leakage before release. Here the decisive dependency is: Logical deletion and physical media sanitization are distinct processes whose evidence depends on the service and control boundary. Use the specified intervention: Confirm the provider's sanitization and media-handling assurances and use controlled cryptographic erase where appropriate.
B:Classify prompt and output artifacts, redact sensitive fields and apply retention, access and regional restrictions consistent with source data. Here the decisive dependency is: Logical deletion and physical media sanitization are distinct processes whose evidence depends on the service and control boundary. Use the specified intervention: Confirm the provider's sanitization and media-handling assurances and use controlled cryptographic erase where appropriate.
C:Plan long-term archive format, key custody, integrity checks and periodic restoration tests before decommissioning dependencies. Here the decisive dependency is: Logical deletion and physical media sanitization are distinct processes whose evidence depends on the service and control boundary. Use the specified intervention: Confirm the provider's sanitization and media-handling assurances and use controlled cryptographic erase where appropriate.
Question 250
Which end-to-end operational sequence best controls risk from detection through acceptance? A regulated company wants to exit its cloud vendor, but its data export is portable only as an undocumented proprietary database image. Select the ONE best answer.
- Investigate: Inspect participating tenants, operator responsibilities, membership criteria, and cloud network exposure and tenancy controls. Remediate: Classify the deployment as a community cloud when access and governance are restricted to organizations with shared concerns. Verify: Architecture documentation identifies membership, shared governance, data separation, and appropriate cloud deployment classification.
- Investigate: Inspect ground-truth labels, false-positive rates, current features, drift alerts, and sample analyst adjudications. Remediate: Validate training-data quality, current drift metrics, outcome feedback, and human review before retuning automated response thresholds. Verify: A holdout test reduces false positives without losing agreed detection recall and records decisions for disputed alerts.
- Investigate: Review data schemas, export formats, egress costs, API dependencies, and the target restore test results. Remediate: Specify an exit architecture with documented open export formats, data mapping, restore tests, and contractual assistance before migration. Verify: An independent pilot imports the exported data into the alternate platform and passes an integrity and functional validation.
- Investigate: Inspect the service-level responsibility matrix, VM administrative privileges, runtime service boundaries, and patch-history ownership. Remediate: Classify the hosting service by the actual operating-system and runtime responsibilities before choosing the patch and monitoring plan. Verify: Documented control ownership matches the actual service contract and guest patch responsibility is assigned to the correct party.
Explanation
Correct Answer
C:Portability and reversibility require usable data and configuration outside a provider, not just nominal permission to download bytes. Review data schemas, export formats, egress costs, API dependencies, and the target restore test results.
Incorrect Answers
A:Inspect participating tenants, operator responsibilities, membership criteria, and cloud network exposure and tenancy controls. Remediate: Classify the deployment as a community cloud when access and governance are restricted to organizations with shared concerns. Verify: Architecture documentation identifies membership, shared governance, data separation, and appropriate cloud deployment classification. These steps address a different operational problem. They do not connect diagnosis, correction and acceptance criteria for the current requirement. The required change is: Specify an exit architecture with documented open export formats, data mapping, restore tests, and contractual assistance before migration.
B:Inspect ground-truth labels, false-positive rates, current features, drift alerts, and sample analyst adjudications. Remediate: Validate training-data quality, current drift metrics, outcome feedback, and human review before retuning automated response thresholds. Verify: A holdout test reduces false positives without losing agreed detection recall and records decisions for disputed alerts. These steps address a different operational problem. They do not connect diagnosis, correction and acceptance criteria for the current requirement. The required change is: Specify an exit architecture with documented open export formats, data mapping, restore tests, and contractual assistance before migration.
D:Inspect the service-level responsibility matrix, VM administrative privileges, runtime service boundaries, and patch-history ownership. Remediate: Classify the hosting service by the actual operating-system and runtime responsibilities before choosing the patch and monitoring plan. Verify: Documented control ownership matches the actual service contract and guest patch responsibility is assigned to the correct party. These steps address a different operational problem. They do not connect diagnosis, correction and acceptance criteria for the current requirement. The required change is: Specify an exit architecture with documented open export formats, data mapping, restore tests, and contractual assistance before migration.