AWS Certified AI Practitioner Practice Test

A company makes forecasts each quarter to decide how to optimize operations to meet expected demand. The company uses ML models to make these forecasts. An AI practitioner is writing a report about the trained ML models to provide transparency and explainability to company stakeholders. What should the AI practitioner include in the report to meet the transparency and explainability requirements?

A law firm wants to build an AI application by using large language models (LLMs). The application will read legal documents and extract key points from the documents. Which solution meets these requirements?

A company wants to use a pre-trained generative AI model to generate marketing content. The company needs to ensure the generated text follows its specific brand voice and tone without retraining the model. Which technique should the company use?

A company is building a Retrieval Augmented Generation (RAG) application on Amazon Bedrock. Which AWS service is most appropriate for storing and querying vector embeddings used for retrieval?

A financial services company is deploying a generative AI assistant that must not leak personally identifiable information (PII) in its responses. Which Amazon Bedrock feature directly addresses this requirement?

A data science team is choosing between supervised and unsupervised learning to detect anomalous credit card transactions. The dataset has millions of transactions but only a tiny fraction are labeled as fraudulent. Which approach is most appropriate?

Which statement best describes a foundation model?

A developer wants to give an LLM access to up-to-date company knowledge stored in internal documents without retraining the model. Which approach should the developer use?

An AI team notices that their model performs significantly worse for one demographic group compared to others. Which responsible AI dimension does this issue most directly relate to?

What is the primary purpose of the temperature parameter when calling a generative LLM?

A company wants to deploy a customer-support chatbot that answers from a curated knowledge base. The team wants a managed solution that handles document ingestion, embedding, and retrieval automatically. Which AWS service is the best fit?

A regulated company needs all data sent to and from a foundation model to remain inside its private network and never traverse the public internet. Which AWS networking feature should it use?

A model achieves 99% accuracy on training data but only 70% accuracy on unseen test data. What is the most likely cause?

Which of the following is the most appropriate metric for evaluating a question-answering system that should produce factually correct responses?

A company plans to publish a model card for its production ML model. Which of the following is the primary purpose of a model card?

Which of these tasks is best solved by a generative AI model rather than a traditional discriminative ML model?

Which AWS service helps track and govern ML model lifecycle events, including who trained a model, with what data, and how it has been used, for audit purposes?

A company wants to lower the cost of running a foundation model that serves a steady, predictable workload at high volume. Which Amazon Bedrock pricing option fits best?

An ML team wants to split its labeled dataset before training. Which split strategy is most appropriate to estimate how well a model will generalize to unseen data?

Which of the following best describes 'hallucination' in the context of generative AI?