Convolutional neural networks (CNNs) are commonly used in generating deepfake videos, where a person's likeness is realistically mapped and animated to create fake but convincing audiovisual content.

Basic Concept: AI chatbots are designed with safety guidelines and content policies that prevent them from generating harmful, offensive, or inappropriate content. When users find ways to bypass these restrictions through crafted prompts, they have " jailbroken " the model. CompTIA SecAI+ Study Guide covers jailbreaking as a key AI vulnerability category.
Why C is Correct: Jailbreaking is the process of using cleverly crafted prompts to bypass an AI model ' s built- in safety restrictions, content policies, and behavioral guardrails, causing it to produce outputs it was designed to refuse. The scenario describes a chatbot that was designed for automobile configuration assistance but is producing offensive responses following customer prompts, indicating that customers have successfully prompted the model to bypass its safety constraints and generate prohibited content.
Why A is Wrong: Model skewing refers to attacks or biases that cause a model to favor certain outputs or perspectives systematically over time, often through data manipulation. It describes a gradual distortion of model behavior, not a direct user-prompted bypass of safety restrictions in a single interaction.
Why B is Wrong: Model theft involves extracting or replicating a proprietary model ' s functionality or architecture through repeated queries. It is an intellectual property attack aimed at stealing the model ' s knowledge, not an attack that causes the model to produce offensive content.
Why D is Wrong: Insecure output handling occurs when an application fails to properly validate or sanitize AI-generated outputs before using them in ways that could cause harm such as passing AI output directly to a system command or database query. It describes a developer implementation vulnerability, not the act of a user prompting a model to bypass its safety constraints.

Consistency ensures that an AI system applies rules and produces results in a uniform manner across all cases. This principle directly addresses concerns about bias and irregular outcomes in the plagiarism detection service.

Option B is correct because the defect is a regression: a security behavior that should require reauthentication was previously corrected but reappeared after a software update. Automated AI-assisted regression testing in the CI/CD pipeline can repeatedly verify that protected account-balance functions still demand reauthentication before a release is promoted. The test should reproduce the prior defect, fail the build when authentication is bypassed, and preserve the case as a permanent security test. Option A adds an approval step, but an approver may not detect the reintroduced behavior without an executable test. Option C improves release scheduling and traceability but does not prove that security controls still function. Option D uses SAST to inspect source code for known weakness patterns; it may identify some authentication flaws, but it does not reliably validate the end-to-end runtime behavior described here. Regression testing is the direct control for preventing recurrence. The NIST AI Risk Management Framework emphasizes testing AI systems before deployment and regularly during operation, supporting automated security validation within the delivery pipeline.

Basic Concept: Input validation is a fundamental security principle that checks incoming data against expected criteria before processing it. For AI systems, this requires a mechanism capable of inspecting the semantic content and structure of inputs - not just their volume or format. CompTIA SecAI+ Study Guide identifies prompt firewalls as the primary input validation control for AI systems.
Why A is Correct: A prompt firewall validates incoming inputs by inspecting their content against security policies, detecting malicious patterns such as injection strings or jailbreaking attempts, enforcing structural rules, and blocking non-compliant inputs before they reach the AI model. Unlike network firewalls that operate on packet headers, a prompt firewall understands the semantic content of AI prompts, making it the appropriate input validation mechanism for AI systems.
Why B is Wrong: Rate limits control how frequently inputs are submitted, not what those inputs contain. A malicious prompt submitted within rate limits will not be detected or blocked - rate limiting does not validate the content or intent of individual inputs.
Why C is Wrong: Token limits cap the maximum length of inputs and outputs in terms of tokens. While this can prevent excessively long inputs from being processed, it does not inspect input content for malicious patterns or validate that inputs conform to policy requirements.
Why D is Wrong: Input quantity is a generic term that might refer to limiting the number or size of inputs.
Like token limits and rate limits, quantity controls do not validate the content of inputs for security compliance or detect malicious prompt patterns.