AWS Certified AI Practitioner practice questions
218 free questions with answers and explanations.
- 1.A research team is experimenting with a new foundation model that exhibits 'emergent abilities.' They observe that as the model's size and training data scale up significantly, it unexpectedly gains capabilities like complex reasoning, problem-solving, and instruction following, which were not explicitly trained for or present in smaller versions of the model. What does this phenomenon imply about the development and understanding of foundation models?Foundation Models
- 2.A financial institution is developing an AI system to detect fraudulent transactions by analyzing patterns in transaction data. They initially consider using a pre-trained foundation model. However, they realize that the unique, highly structured, and numerical nature of transaction data, alongside strict privacy regulations, makes direct application challenging. Which characteristic of foundation models presents a significant challenge in this scenario compared to traditional machine learning models for structured numerical data?Foundation Models
- 3.A research team is developing an AI model to detect early signs of a rare disease. Due to the extreme scarcity of positive cases, the training dataset is heavily imbalanced, with very few examples of the disease. Simply maximizing overall accuracy results in a model that almost always predicts 'no disease.' To build a Responsible AI system, which metric should they primarily focus on to ensure the model can effectively identify positive cases?Responsible AI
- 4.A company is designing an AI system that processes highly sensitive personal data. To comply with data protection regulations like GDPR, they implement techniques such as anonymization, pseudonymization, and secure data storage. Which Responsible AI concept are these practices primarily intended to uphold?Responsible AI
- 5.A government agency is using an AI system to process citizen requests. They are concerned that the AI might make decisions that are difficult to justify or explain to the public, leading to a lack of trust. To mitigate this, they plan to use techniques like LIME (Local Interpretable Model-agnostic Explanations) and SHAP (SHapley Additive exPlanations). Which Responsible AI concept are these techniques primarily designed to address?Responsible AI
- 6.A financial institution is developing an AI model to assess loan applications. To ensure fairness, the data scientists are particularly concerned about disparate impact on protected demographic groups, even if the model's performance metrics (accuracy, precision, recall) are high overall. Which of the following Responsible AI concepts is primarily being addressed in this scenario?Responsible AI
- 7.A team is developing an AI system for predictive policing. They are conducting a thorough assessment to identify potential negative societal impacts, ethical dilemmas, and risks of discrimination before deploying the system. This proactive evaluation process is known as a(n):Responsible AI
- 8.An AI engineer is evaluating a pre-trained foundation model for a task that involves question answering over a large, dynamic dataset of company policies. The model performs well on general knowledge questions but struggles with specific, up-to-date policy details, sometimes providing outdated or incorrect information. The engineer wants to improve the model's accuracy on this specific, evolving dataset without performing extensive and frequent retraining. Which approach would be most effective?Foundation Models
- 9.A development team is building a content moderation system that needs to identify and flag inappropriate images and videos automatically. They are considering using a foundation model to accelerate development. Which characteristic of foundation models makes them particularly suitable for this task, especially given limited labeled data for specific inappropriate content categories?Foundation Models
- 10.A software development company is migrating its internal documentation to a knowledge base powered by a foundation model. They want to ensure that the model can understand and accurately respond to queries about their proprietary software features, which are unique and not covered by publicly available training data. What is the most effective strategy to adapt a pre-trained foundation model for this highly specialized domain?Foundation Models
- 11.A company is developing a new customer service chatbot and needs to select a foundation model. They are evaluating two models: Model A, which is a proprietary model with a very large parameter count and extensive pre-training; and Model B, which is an open-source model with a smaller parameter count but has been fine-tuned on a public customer service dataset. The company has a tight budget and limited GPU resources. Which factor is MOST critical in deciding between these two models given the company's constraints?Foundation Models
- 12.A product manager is evaluating the use of foundation models for a new feature that automatically generates marketing copy for various product categories. They notice that while the model can produce grammatically correct and fluent text, it sometimes struggles to accurately capture the specific brand voice and tone, or include unique selling propositions (USPs) that are critical to the company's messaging. Which concept best describes this challenge?Foundation Models
- 13.An AI solutions architect is evaluating different approaches for a client who wants to build a custom intelligent assistant for their specialized legal firm. The assistant needs to answer complex legal questions based on the firm's private document repository. The architect decides to use a pre-trained Large Language Model (LLM) and augment it with the firm's data. Which technique is most appropriate for integrating the firm's private legal documents into the LLM's knowledge base without retraining the entire model?Foundation Models
- 14.A credit scoring AI model is found to consistently give lower scores to individuals residing in certain zip codes, even when controlling for other financial factors. This leads to higher loan rejection rates for these individuals. This situation exemplifies which type of bias in AI?Responsible AI
- 15.A developer is using a foundation model for code generation. When providing a prompt like 'Write a Python function to sort a list of integers,' the model consistently generates code that uses a bubble sort algorithm, even when more efficient algorithms (like quicksort or mergesort) would be better. This behavior is likely due to the model's pre-training data containing a disproportionately higher number of bubble sort examples. This scenario is an example of which ethical concern related to foundation models?Foundation Models
- 16.A startup is developing an AI-powered chatbot for mental health support. Given the sensitive nature of the data and the potential for harm, they are establishing a framework to clearly define who is responsible for model errors, data breaches, or unintended negative consequences. Which Responsible AI concept are they primarily focusing on?Responsible AI
- 17.A healthcare provider is implementing an AI system to assist with disease diagnosis. During development, it's discovered that the model performs significantly worse on data from a particular ethnic minority group due to underrepresentation in the training dataset. Which of the following is the most appropriate best practice to address this issue?Responsible AI
- 18.An e-commerce company uses an AI system to personalize product recommendations. A customer complains that they are consistently shown products completely unrelated to their browsing history or past purchases. The company wants to understand why the AI made these specific recommendations for this customer. Which Responsible AI concept are they trying to implement?Responsible AI
- 19.A data scientist is exploring different foundation models for a new project that requires generating human-like text responses for a chatbot. The project needs a model capable of understanding context and producing coherent, grammatically correct, and relevant replies. Which type of foundation model is best suited for this task?Foundation Models
- 20.A healthcare provider wants to use a foundation model to assist doctors in diagnosing rare diseases by analyzing patient medical images and clinical notes. Due to the highly sensitive nature of patient data and strict compliance requirements (e.g., HIPAA), they cannot send raw patient data to a public cloud API for inference. Which deployment strategy for the foundation model would best address these stringent data privacy and compliance needs?Foundation Models
- 21.A startup is developing an application that uses a foundation model to summarize long technical reports. They are encountering issues where the model sometimes generates plausible-sounding but factually incorrect information, a phenomenon known as 'hallucination.' Which fundamental limitation of foundation models contributes most directly to this issue?Foundation Models
- 22.An AI-powered content moderation system is being developed for a social media platform. To prevent the system from being easily manipulated by malicious actors attempting to bypass its filters with slight modifications to harmful content, the development team is employing techniques like adversarial training. Which Responsible AI concept are they prioritizing?Responsible AI
- 23.A company is deploying an AI system for automated hiring. To adhere to Responsible AI best practices, they establish a clear process for human oversight, including review points where a human can override the AI's decision and mechanisms for appealing decisions. Which aspect of Responsible AI is this practice primarily addressing?Responsible AI
- 24.A research institution is developing a foundation model for scientific discovery, specifically to analyze complex genomic data. They are considering using a multi-modal foundation model. What is the primary advantage of a multi-modal foundation model in this context compared to a single-modality model (e.g., text-only or image-only)?Foundation Models
- 25.A financial services company processes millions of customer documents daily, including scanned forms and PDFs. They need to extract specific data fields, such as account numbers, names, and transaction amounts, from these documents for regulatory compliance and automated processing. The documents often have varying layouts and are sometimes handwritten. Which AWS service should they use?AWS Services for AI/ML and Generative AI
- 26.A pharmaceutical company is developing a new drug discovery platform that involves training highly sensitive machine learning models on patient genomic data. To meet stringent regulatory and security compliance, they need to ensure that all data used for training and inference is encrypted at rest and in transit, and that the encryption keys are managed and controlled by the company, not solely by AWS. Which AWS service combination provides the necessary encryption and key management capabilities for this scenario?AWS Services for AI/ML and Generative AI
- 27.A global technology company is developing an AI-powered translation service. Due to its worldwide reach, the company wants to ensure that the AI's output is culturally appropriate and avoids offense in diverse linguistic and cultural contexts. This goes beyond mere linguistic accuracy to address nuances in social norms and values. Which Responsible AI consideration is most relevant here?Responsible AI
- 28.A financial institution is developing an AI model to detect fraudulent transactions. The model must be highly accurate and reliable, as false positives could inconvenience legitimate customers, and false negatives could lead to significant financial losses. To ensure the model performs consistently even with novel or slightly altered fraudulent patterns, the development team is rigorously testing its resilience against various adversarial inputs and data shifts. Which Responsible AI concept is being prioritized here?Responsible AI
- 29.A team is fine-tuning a pre-trained foundation model for a highly specialized legal domain. They have a relatively small dataset of annotated legal documents. They want to adapt the model to understand the nuances of legal language without losing the broad general knowledge it gained during its initial massive pre-training. Which fine-tuning strategy is most appropriate to achieve this balance?Foundation Models
- 30.A startup is building a new e-commerce platform and wants to provide personalized product recommendations to its users in real time. They need a fully managed machine learning service that can leverage historical user behavior data (e.g., clicks, purchases) to generate recommendations without requiring deep machine learning expertise. Which AWS service should they choose?AWS Services for AI/ML and Generative AI
- 31.A company is experimenting with a generative AI model to create marketing copy. They want to use a foundation model (FM) offered by AWS, but also need the ability to fine-tune it with their proprietary brand guidelines and product information to generate more specific and on-brand content. Which AWS generative AI service allows them to do this?AWS Services for AI/ML and Generative AI
- 32.A government agency is using an AI system to process citizen requests for social benefits. They are concerned that the system might make decisions that are difficult to understand or justify to citizens, leading to distrust. To mitigate this, they plan to implement methods that can explain the reasoning behind the AI's recommendations in simple, human-readable language, especially for rejected applications. Which Responsible AI concept are they prioritizing?Responsible AI
- 33.An e-commerce company uses an AI system to personalize product recommendations. Customers have reported that the recommendations sometimes feel repetitive or do not align with their current interests despite recent purchases. To improve the user experience and maintain trust, the company wants to allow customers to understand why a particular product was recommended. Which Responsible AI concept is the company trying to implement?Responsible AI
- 34.A company is migrating its on-premises document processing workflow to AWS. They frequently deal with scanned invoices and receipts, needing to extract structured data such as vendor names, invoice numbers, and line items, even from poorly formatted documents. Which AWS service is purpose-built for this task?AWS Services for AI/ML and Generative AI
- 35.A global media company uses AWS for its content delivery network and streaming services. They are exploring generative AI to automatically create short promotional video clips from longer content. Due to regulatory requirements and intellectual property concerns, all models trained and used for content generation must reside exclusively within specific geographic regions (e.g., Europe for European content) and never leave those boundaries. Which AWS mechanism ensures that the generative AI services operate within the specified geographic regions and that data residency requirements are met?AWS Services for AI/ML and Generative AI
- 36.A startup is developing an application that uses a foundation model to summarize long technical documents. After initial testing, users report that the summaries occasionally contain plausible-sounding but factually incorrect information not present in the original document. Which term best describes this phenomenon?Foundation Models
- 37.A content creation company wants to generate unique blog posts and marketing copy based on short prompts provided by their marketing team. They need a model that can produce coherent, contextually relevant, and stylistically appropriate long-form text. Which type of foundation model is best suited for this specific application?Foundation Models
- 38.A large e-commerce company wants to implement a foundation model to improve its customer service chatbot. The chatbot needs to handle a wide variety of customer inquiries, from product information to order tracking and returns, and provide coherent, contextually appropriate responses. Which key characteristic of foundation models makes them particularly suitable for this general-purpose conversational AI task?Foundation Models
- 39.A credit scoring AI model is found to consistently give lower scores to individuals residing in certain low-income neighborhoods, regardless of their individual financial history. This happens because the model implicitly learns correlations between zip codes (which are proxies for income) and creditworthiness from historical biased data. This phenomenon is best described as:Responsible AI
- 40.A media company wants to automatically generate summaries of news articles and detect key entities like people, organizations, and locations within the text. They also need to understand the overall sentiment of the articles. Which AWS AI service would be most suitable for these tasks?AWS Services for AI/ML and Generative AI
- 41.A healthcare organization is developing an AI model to assist with patient diagnosis. To ensure the model is fair and does not inadvertently discriminate against certain demographic groups, the data scientists are evaluating the model's performance across different patient populations (e.g., age, gender, ethnicity). Which Responsible AI concept are they primarily addressing?Responsible AI
- 42.A human resources department is implementing an AI-powered resume screening tool to streamline the hiring process. To prevent unintentional discrimination and ensure diverse candidate pools, the HR team plans to regularly audit the AI's decisions, compare its shortlists against human-generated ones, and provide a mechanism for candidates to appeal AI-driven rejections. This oversight process aims to ensure that humans remain ultimately in charge of critical decisions. Which Responsible AI concept is primarily being applied?Responsible AI
- 43.A data scientist is evaluating various foundation models for a project that involves generating creative marketing copy. The client requires the model to produce diverse and imaginative text based on short prompts. Which characteristic of foundation models is most relevant for this use case?Foundation Models
- 44.A data science team is evaluating several pre-trained foundation models for a complex natural language processing task that involves understanding nuanced language and performing multi-step reasoning, such as answering complex questions that require inferring information not explicitly stated. They observe that some models, when scaled up significantly in parameters and training data, suddenly become capable of solving tasks they were not explicitly trained for, like performing basic arithmetic or translating between languages with high accuracy. What phenomenon are they observing?Foundation Models
- 45.A financial institution is developing an AI model to predict credit default. The data scientists observe that the model consistently performs worse for applicants from a specific demographic group, leading to higher false positive rates for default prediction within that group. Which Responsible AI concept is primarily being violated?Responsible AI
- 46.A development team is building a content moderation system that needs to identify and flag inappropriate content across various social media platforms. The system must adapt to new types of inappropriate content quickly without requiring extensive retraining for every new category. Which characteristic of foundation models makes them particularly suitable for this scenario?Foundation Models
- 47.A financial institution is developing an application that analyzes transaction data to detect fraudulent activities in real-time. They need a service that can quickly process continuous streams of data and prepare it for machine learning models. Which AWS service is best suited for this requirement?AWS Services for AI/ML and Generative AI
- 48.An AI research lab is developing a new foundation model with billions of parameters. During the training phase, they observe that the model's performance on the training dataset is exceptionally high, but its performance on unseen validation data is significantly lower. This indicates a problem where the model has learned the training data too well, including its noise and specific patterns, rather than general principles. What is this phenomenon called?Foundation Models
- 49.A startup is developing a new mobile application that allows users to upload photos, and they need to automatically identify objects, scenes, and faces within these images for content moderation and tagging. They also want to detect any inappropriate content to ensure a safe user experience. Which AWS service would provide these capabilities with minimal machine learning expertise required?AWS Services for AI/ML and Generative AI
- 50.An AI-powered content moderation system is being developed for a social media platform. To ensure Responsible AI, the system must be able to maintain its performance and accuracy even when faced with subtle variations in input data, such as slightly rephrased offensive sentences or images with minor alterations designed to bypass detection. Which Responsible AI concept is most critical to address this requirement?Responsible AI