AI Concepts CERTIFIED-TECHNOLOGY-SOLUTIONS-PROFESSIONAL Practice Questions (Page 5)

Questions 21–25

1. 21foundation · easy

   Which security measure is essential to protect sensitive data used in AI training from unauthorized access?

   • AUsing a parallel file system
   • BEncryption at rest and in transit
   • CUsing a public cloud deployment
   • DIncreasing the number of GPUs
   Reveal Solution Discussion Select an answer first
2. 22foundation · easy

   Which AI workload is primarily responsible for using a trained model to make predictions on new, unseen data?

   • ATraining
   • BInference
   • CData preparation
   • DModel validation
   Reveal Solution Discussion Select an answer first
4. 23application · medium

   A multinational corporation is deploying an AI model that uses personal data from customers in multiple countries. The legal team requires that the model's access be restricted to authorized personnel and that all data handling comply with regional privacy regulations. Which combination of measures should the IT team implement?

   • AEncrypt data at rest and in transit, and implement role-based access control (RBAC) for the model
   • BUse a public cloud without specific region selection
   • CStore all data in a single on-premises data center
   • DDisable all logging and monitoring to protect privacy
   Reveal Solution Discussion Select an answer first
5. 24foundation · easy

   Which statement correctly describes the relationship between artificial intelligence (AI), machine learning (ML), and deep learning (DL)?

   • AML is a subset of AI, and DL is a subset of ML that uses multi-layered neural networks.
   • BAI, ML, and DL are three distinct and unrelated approaches to building intelligent systems.
   • CAI is a subset of ML that focuses on neural networks with many layers.
   • DDL and ML are independent fields that together form AI.
   Reveal Solution Discussion Select an answer first
6. 25application · medium

   A data science team is training a model on a large dataset. They notice that the GPUs are often idle while data is being loaded from storage. They want to improve GPU utilization. Which technique would directly help?

   • AIncreasing the batch size
   • BReducing the number of GPUs
   • CUsing a slower storage system to reduce I/O errors
   • DData caching to keep frequently used data in faster storage
   Reveal Solution Discussion Select an answer first