An enterprise has approved multiple pilots and early-stage AI use cases across different functions. Adoption teams are still evaluating which workflows deliver consistent productivity and quality improvements. At this stage, leadership wants to avoid creating administrative overhead that could slow experimentation or discourage participation. Financial monitoring is being handled centrally while usage patterns and business impact are still being analyzed, and individual business units are not yet being asked to account for their own consumption. Which cost accountability approach is being applied in this phase?

• A. Team-based budgeting
• B. Showback model
• C. Chargeback model
• D. Centralized model

During an AI initiative review, a delivery team reports that a predictive model is underperforming despite using datasets that already meet established quality, completeness, and consistency standards. The data has been sourced and validated, and no changes to model design or additional data acquisition are planned at this stage. Analysis indicates that existing data fields do not sufficiently reflect higher-level business behavior needed for learning. As part of AI operations oversight, you are asked to identify which data preparation activity should be applied next to address this issue. Which activity within the Data Collection and Preparation phase directly supports improving how existing data is represented for model learning?

• A. Creating meaningful variables from existing data
• B. Applying ground truth labels to records
• C. Extracting raw data from source systems
• D. Dividing data into training, validation, and test sets

Julianne Moore, Lead AI Systems Architect, is conducting an investigation on a facial recognition access system that recently failed a security audit. The audit team demonstrated that by wearing a specifically crafted pair of noisy pattern eyeglasses, an unauthorized user could consistently trick the system into identifying them as the CEO. Julianne confirms that the system's source code is intact and the original database of face images used to train the model was verified as clean and unaltered. Julianne must categorize this vulnerability in her report to the CISO. Which AI-specific security threat characterizes the method used to bypass the system's identification controls?

• A. Data Poisoning
• B. Model Theft
• C. Adversarial Attacks
• D. Prompt Injection

Audrey is the Chief Legal Officer for a multinational software corporation. As the company prepares to launch a high-risk AI application globally, Audrey advises the board to prioritize a specific regional framework as the foundation for their internal compliance program. She argues that because this framework represents the most comprehensive, risk-based standard currently in existence, adhering to it will likely satisfy the core requirements of other regional regulations the company must navigate. Which specific regulatory framework is Audrey referencing as the most comprehensive standard influencing global compliance?

• A. Singapore FEAT
• B. EU AI Act
• C. OECD AI Principles
• D. NIST AI RMF

A retail organization is preparing historical sales data for retraining a demand-forecasting model. Initial checks confirm that all required fields are populated, values reflect real operational records, and duplicate entries have already been removed. However, during automated pipeline execution, multiple transformation steps fail unpredictably across different batches. Investigation shows that some records violate predefined structural constraints used by downstream processing logic, even though the underlying business values appear reasonable. Before retraining proceeds, the Data Engineering Lead pauses the pipeline to address the underlying issue to ensure stable execution. Which data quality dimension is primarily impacted in this scenario?

• A. Alignment with real-world conditions
• B. Presence of required data elements
• C. Conformance to defined rules and constraints
• D. Availability of up-to-date records