Evaluate Initial AI Feasibility PMI-CPMAI Practice Questions (Page 2)

Questions 6–10

1. 6foundation · easy

   What is the purpose of comparing AI solutions with traditional non-AI alternatives?

   • ATo assess the quality of available data
   • BTo determine if AI is the best approach considering benefits, costs, risks, and feasibility
   • CTo estimate the computational resources for AI training
   • DTo evaluate the organization's readiness for AI
   Reveal Solution Discussion Select an answer first
2. 7foundation · easy

   What does computational resource estimation primarily determine?

   • AThe quality of the training data
   • BThe hardware, software, and time needed for model training and inference
   • CThe accuracy of the model on test data
   • DThe business value of the AI solution
   Reveal Solution Discussion Select an answer first
4. 8expert · hard

   A media company wants to use AI to automatically generate video captions. They plan to train a large speech recognition model on a massive dataset of audio files. The company has a limited budget and must decide between renting cloud GPUs or building an on-premises GPU cluster. The company's IT team has experience with cloud services but no experience with on-premises GPU clusters. What is the most important factor in evaluating feasibility?

   • AThe need to integrate the caption generation with the company's video editing software.
   • BThe availability of open-source speech recognition models that can be fine-tuned instead of training from scratch.
   • CThe need to hire a team of data scientists to label the audio data.
   • DThe total cost of cloud GPU rental versus the cost of building and maintaining an on-premises cluster, including the team's ability to manage it.
   Reveal Solution Discussion Select an answer first
5. 9expert · hard

   A bank wants to use AI to approve credit card applications. They have a large dataset of historical applications, but the data is biased against certain demographic groups. The bank is subject to strict regulatory requirements for fair lending. The bank's IT team is experienced with traditional credit scoring models. What should the bank do to evaluate initial AI feasibility?

   • AAssess whether the bias in the data can be mitigated to meet regulatory requirements, and compare the AI model's performance against the existing credit scoring model.
   • BHire a third-party AI vendor to build the model, as they will handle regulatory compliance.
   • CProceed with AI because it is more accurate than traditional models, regardless of bias.
   • DFocus only on cleaning the data to remove bias, without considering the existing model.
   Reveal Solution Discussion Select an answer first
6. 10application · medium

   A hospital wants to use AI to predict patient readmission risk. They have electronic health records for the last five years, but many records have missing discharge summaries, and some fields are entered inconsistently by different departments. What is the most critical feasibility concern to address first?

   • AWhether the hospital has enough GPUs to train a deep learning model on the full dataset.
   • BWhether the missing and inconsistent data can be cleaned or imputed to a level that supports reliable model training.
   • CWhether the hospital's IT department can integrate the AI model with the existing patient management system.
   • DWhether the hospital has a data protection officer to ensure compliance with health data regulations.
   Reveal Solution Discussion Select an answer first