AWS Certified AI Practitioner (AIF-C01) — All Questions

What most clearly distinguishes generative AI from traditional discriminative machine learning?

• a.Generative AI is always faster and cheaper to run than every other type of machine learning model
• b.Generative AI creates new content such as text or images, not just labels for existing data✓
• c.Generative AI needs no training data whatsoever and can function from nothing but a short prompt
• d.Generative AI is able only to sort each of its inputs into a set of predefined fixed categories

Generative models produce new artifacts (text, images, audio, code) that resemble their training data. Discriminative models instead label or score existing inputs. Generative AI still needs training data and is often compute-intensive.

In the context of large language models, what is a token?

• a.A numeric confidence score that rates how truthful each given answer is likely
• b.A complete source document stored in the model's memory
• c.A chunk of text, such as a word or sub-word piece, processed as one unit✓
• d.A security credential that authorizes an API call

LLMs break text into tokens (words or sub-word fragments) and process them as units. Token counts drive context limits and pricing. This is unrelated to auth tokens, which are security credentials.

What is a foundation model?

• a.A specialized database index that speeds up structured queries
• b.A rule-based engine that only stores hand-written if-then statements from experts
• c.A small model trained entirely from scratch for one single narrowly defined task
• d.A large model pre-trained on broad data that adapts to many downstream tasks✓

Foundation models are large models pre-trained on massive, broad datasets, giving them general capabilities that can be adapted (via prompting or fine-tuning) to many tasks, rather than being built for a single narrow purpose.

Which AWS service provides access to foundation models from multiple providers through a single serverless API?

• a.Amazon Comprehend, a natural language service that finds named entities in text
• b.Amazon Polly, a service that turns text into speech
• c.Amazon Textract, a service that extracts text from files
• d.Amazon Bedrock, a unified serverless gateway to many foundation models✓

Amazon Bedrock offers a unified serverless API to foundation models from providers such as Anthropic, Meta, Cohere, and Amazon. Comprehend, Textract, and Polly are single-purpose AI services, not FM access layers.

What does it mean when a large language model 'hallucinates'?

• a.It slows down noticeably because the input prompt it received was simply far too long
• b.It stores the user's private personal data permanently without ever obtaining any consent
• c.It produces fluent, confident output that is actually factually incorrect or fabricated✓
• d.It refuses outright to answer any question that happens to involve numbers or arithmetic

A hallucination is generated content that sounds plausible and confident but is factually wrong or invented. It is a core reliability risk of LLMs and a reason to ground outputs in verified sources and to review them.

A company wants a foundation model to answer questions using its own internal documents without retraining the model. Which technique fits best?

• a.Removing the guiding system prompt so the model answers freely
• b.Retrieval Augmented Generation, which fetches relevant documents and adds them to the prompt✓
• c.Training an entirely brand-new foundation model completely from scratch on those same documents
• d.Increasing the model's temperature setting all the way up to a very high randomness value

RAG retrieves relevant passages from a knowledge source at query time and injects them into the prompt, grounding answers in current, proprietary data without the cost of retraining. Higher temperature only adds randomness.

What is the most likely effect of setting a very high temperature value when generating text from an LLM?

• a.The output turns strictly deterministic and comes out identical on every single run without variation
• b.The model begins to automatically fact-check every sentence it writes
• c.The model quietly ignores the entire prompt and returns an empty response
• d.The generated output becomes more random, more varied, and generally more creative✓

Temperature controls randomness in token sampling. Higher temperature yields more diverse, creative, less predictable text; lower temperature makes output more focused and repeatable. It does not verify facts.

Which of the following is a realistic and appropriate use case for generative AI?

• a.Serving as the single, completely unreviewed final authority for patient medical diagnoses
• b.Drafting a first version of marketing copy that a human then reviews and edits✓
• c.Fully replacing every relational and object data storage system across the whole organization
• d.Guaranteeing perfectly accurate binding legal rulings with absolutely no human oversight at all

Generative AI is well suited to drafting content that humans review, such as marketing copy. In high-stakes domains like medicine or law it should assist experts, not replace their judgment, because of hallucination and accountability risks.

What is an embedding in the context of generative AI and semantic search?

• a.A numeric vector representation of data that captures its underlying semantic meaning✓
• b.The specific physical AWS Region and data-center server location where a model is hosted
• c.A hard-coded list of banned words the model must avoid
• d.A heavily compressed image file used to reduce storage cost

Embeddings map text (or images, audio) into numeric vectors so that items with similar meaning sit close together in vector space. They power semantic search, clustering, and RAG retrieval.

How does a general-purpose foundation model differ from a narrow, task-specific traditional ML model?

• a.It can handle only one single narrow task, unlike the more flexible traditional model
• b.It requires no training data , while a traditional model needs data
• c.It is genuinely identical in scope and differs only in name
• d.It is broadly capable and adaptable to many tasks, not built for a single one✓

Foundation models are trained broadly and adapt to many downstream tasks with prompting or light tuning. Traditional ML models are typically trained for one specific task and do not generalize beyond it.

A developer wants a coding assistant integrated with AWS that can answer questions and help write and debug code. Which service is designed for this?

• a.Amazon Q Developer, the generative AI assistant built for coding and AWS tasks✓
• b.Amazon Forecast, the fully managed service that predicts future numeric values from historical time-series data
• c.Amazon Kendra, the enterprise search service
• d.Amazon Transcribe, the service that converts speech into text

Amazon Q Developer is the generative AI assistant for software development and AWS tasks, helping write, explain, and debug code. Kendra is enterprise search, Transcribe is speech-to-text, and Forecast handles time-series prediction.

Which factor should most influence the choice of one foundation model over another for a given application?

• a.Only the alphabetical order of the model provider's name
• b.Whichever particular model happened to be released most recently, ignoring every other factor
• c.The trade-offs among capability, latency, cost, and fit for the specific task✓
• d.Always picking the model with the very largest parameter count regardless of need

Model selection balances task fit, output quality, latency, and cost. The largest or newest model is not automatically best; a smaller, cheaper model may meet requirements with lower cost and faster responses.

What is a prompt in the context of a large language model?

• a.The input text or instructions given to the model to guide the response it generates✓
• b.The lossless compression algorithm that is used to store the model's weights on disk
• c.The graphics processing unit hardware that actually runs the model's heavy computations
• d.The billing invoice document that is automatically generated after every single API call

A prompt is the input, the instructions, context, and questions, that you provide to steer the model's output. Crafting effective prompts (prompt engineering) strongly shapes the quality and relevance of responses.

What does a large language model's 'context window' refer to?

• a.The maximum amount of text, measured in tokens, the model can consider at once, including prompt and response✓
• b.The physical AWS Region where the model happens to be hosted
• c.The set of safety filters that are applied to the model's output
• d.The graphical console window used to chat with the model

The context window is the token budget the model can attend to in a single request (input plus output). Exceeding it forces truncation. It is about capacity, not hosting location or safety filtering.

The Top P (nucleus sampling) inference parameter controls what?

• a.The maximum number of tokens allowed in the response
• b.The AWS Region that is used to run inference
• c.Sampling from the smallest set of tokens whose cumulative probability reaches a threshold, affecting output diversity✓
• d.Whether the model's weights are updated during the request

Top P restricts sampling to the most probable tokens that together reach a cumulative probability p, balancing diversity and focus. It never updates weights and is separate from the max-tokens length limit.

A team wants to generate original images from written text descriptions. Which type of foundation model do they need?

• a.A text-embeddings model
• b.A speech-to-text model
• c.A tabular regression model
• d.A text-to-image generation model, such as Amazon Titan Image Generator or Stable Diffusion on Amazon Bedrock✓

Creating images from text prompts requires a text-to-image generative model (for example Titan Image Generator or Stable Diffusion). Embeddings models output vectors and speech-to-text handles audio, not image creation.

What is a multimodal foundation model?

• a.A model that can process only plain text
• b.A model that is jointly owned by several companies
• c.A model that is hosted in multiple AWS Regions simultaneously
• d.A model that can process and/or generate more than one type of data, such as text and images together✓

Multimodal models work across multiple data types (for example accepting an image plus a text question). A text-only model is unimodal. 'Multimodal' describes the data types handled, not hosting or ownership.

Which AWS service is a generative AI assistant that answers questions and generates content grounded in a company's own enterprise data and systems?

• a.Amazon Q Business✓
• b.Amazon Rekognition
• c.Amazon Q Developer
• d.Amazon Comprehend

Amazon Q Business is an enterprise generative AI assistant that connects to company data sources to answer questions and generate content. Amazon Q Developer targets coding and AWS development tasks instead.

In Amazon Bedrock, what is 'Amazon Titan'?

• a.Amazon's own family of foundation models available through Amazon Bedrock✓
• b.A managed vector database service
• c.A no-code model-building tool for business analysts
• d.A hardware chip used for training large models

Amazon Titan is Amazon's family of first-party foundation models (for text, embeddings, and images) offered via Bedrock. Amazon's training and inference chips are Trainium and Inferentia, which are separate products.

Why might a large language model be unable to correctly answer a question about an event that happened last week?

• a.The model automatically refuses any question that mentions a date
• b.The temperature parameter was set too low for recent events
• c.The prompt was encrypted before it reached the model
• d.Its knowledge is limited to its training-data cutoff, and it has no inherent access to real-time information✓

An LLM only knows what was in its training data up to a cutoff and cannot see live events unless connected to a tool or retrieval system. This is why techniques like RAG are used to supply current information.

Condensing a long report into a short paragraph of its key points is which generative AI task?

• a.Classification
• b.Clustering
• c.Text summarization✓
• d.Named-entity extraction

Producing a shorter version that preserves the key points is text summarization, a common generative AI use case. Classification assigns labels and clustering groups data, neither of which rewrites the content into a summary.

A team needs to turn text into numeric vectors for a semantic search feature. Which type of model should they use?

• a.An embeddings model, such as Amazon Titan Text Embeddings✓
• b.A text-to-image generation model
• c.A speech synthesis (text-to-speech) voice generation model
• d.A chat model tuned for long conversations

Semantic search relies on embeddings that place similar meanings near each other in vector space, so an embeddings model is the right tool. Chat or image models generate content rather than produce search vectors.

Modern large language models are built primarily on which neural network architecture?

• a.Convolutional neural networks
• b.The transformer architecture✓
• c.Decision trees
• d.Linear regression

Transformers, which use self-attention, are the foundation of modern LLMs and most generative models. CNNs are common in classic computer vision, while decision trees and linear regression are traditional ML methods.

Which offering lets people experiment with and build generative AI apps in a no-code playground based on Amazon Bedrock?

• a.Amazon SageMaker Ground Truth data labeling
• b.Amazon Macie
• c.AWS CloudTrail
• d.PartyRock, an Amazon Bedrock Playground✓

PartyRock is a no-code, hands-on playground built on Amazon Bedrock for experimenting with and sharing generative AI apps. Ground Truth labels data, Macie finds sensitive data, and CloudTrail logs API activity.

A user hides malicious instructions inside input text to trick an LLM into ignoring its original instructions. What is this security risk called?

• a.Overfitting
• b.Data drift
• c.Prompt injection✓
• d.Model hallucination

Prompt injection manipulates a model by smuggling instructions into its input to override intended behavior. Hallucination is fabricated output, and drift or overfitting are training-related issues, not input attacks.

Which is a genuine limitation of generative AI that practitioners must plan for?

• a.It can produce plausible but incorrect (hallucinated) output, and its responses can be non-deterministic✓
• b.It is fundamentally incapable of generating any text
• c.It can never be given instructions of any kind
• d.It always returns identical output for every possible prompt

Generative models can hallucinate and vary their answers across runs, so outputs should be reviewed and, where accuracy matters, grounded in trusted sources. They clearly can generate text and follow instructions.

Building a customer-service virtual assistant that holds natural, open-ended conversations is a common use case for what?

• a.Generative AI large language models✓
• b.A tabular regression model
• c.Optical character recognition
• d.A classical time-series forecasting model

Open-ended conversational assistants are a hallmark generative AI and LLM use case. Regression and forecasting predict numbers, and OCR extracts text from images, none of which conduct free-form dialogue.

A marketing team wants to create original concept artwork from written descriptions. Which generative AI capability fits?

• a.Sentiment analysis
• b.Text-to-image generation✓
• c.Speech-to-text transcription
• d.Anomaly detection

Generating novel images from text prompts is text-to-image generation. Sentiment analysis and anomaly detection analyze existing data, and speech-to-text transcribes audio rather than creating images.

Generating a working function from a natural-language description of what it should do is an example of which generative AI use case?

• a.Code generation✓
• b.Data encryption
• c.Language translation
• d.Image classification

Turning a plain-language request into source code is code generation, a strength of coding-focused models and assistants such as Amazon Q Developer. The other options do not produce code.

Why does token count matter when using a large language model?

• a.Tokens determine how much text fits in the context window and typically drive usage-based pricing✓
• b.Tokens are the encryption keys that secure the model
• c.Tokens are the model's internal learned weights
• d.Token count has no effect on either cost or capacity

Text is processed as tokens; the number of tokens affects both how much content fits in the context window and how much you pay under per-token pricing. Tokens are units of text, not keys or weights.

How are foundation models typically pre-trained?

• a.On a small, fully hand-labeled dataset built for just one narrow task only
• b.On vast amounts of largely unlabeled data using self-supervised learning✓
• c.Without using any data at all
• d.Only on the individual customer's private documents

Foundation models are pre-trained on enormous, broad, mostly unlabeled corpora using self-supervised objectives, which gives them general capabilities. They are later adapted to specific tasks via prompting or fine-tuning.

In semantic search, two pieces of text are judged similar in meaning when what is true of their embeddings?

• a.They contain the exact same characters
• b.They were stored in the same S3 bucket
• c.Their vectors are close together in the embedding space✓
• d.They happen to contain an identical total number of tokens

Embeddings place semantically similar text near each other, so closeness in vector space signals similar meaning even when the wording differs. Matching characters or token counts is not what semantic similarity measures.

A key benefit of Amazon Bedrock is that it is serverless. What does that mean for the customer?

• a.The customer must manage a fleet of GPU servers themselves
• b.The models run only on the customer's on-premises hardware
• c.They can use foundation models through an API without provisioning or managing any underlying infrastructure✓
• d.The service can only be used from inside a physical data center

Serverless means AWS handles the infrastructure; customers simply call foundation models through an API and pay for what they use. There are no servers to provision, patch, or scale on the customer's side.

Which inference parameter limits how long a model's generated response can be?

• a.The maximum tokens (maximum length) setting✓
• b.The Top K setting
• c.The temperature setting
• d.The stop-sequence termination configuration setting

Maximum tokens caps the length of the generated output. Temperature and Top K shape randomness and diversity, and stop sequences end generation at specific text but do not by themselves set an overall length limit.

A company wants a foundation model to reflect its niche industry vocabulary more deeply than prompting can achieve. Which is a valid customization path on Amazon Bedrock?

• a.Increasing the temperature parameter
• b.Switching to a larger context window only
• c.Fine-tuning or continued pre-training the model on the company's own data✓
• d.Adding more stop sequences to each request

To embed specialized knowledge or style beyond what prompts provide, Bedrock supports customization via fine-tuning (labeled data) or continued pre-training (unlabeled domain data). Inference settings like temperature do not add knowledge.

What is a key advantage of generative AI over traditional rule-based automation for content tasks?

• a.It can produce novel, human-like content and adapt to varied inputs without hand-written rules for every case✓
• b.It guarantees perfectly accurate output every single time
• c.It never requires any human review of its output
• d.It eliminates all compute costs entirely

Generative AI flexibly creates new, human-like content and handles many input variations that rigid rules cannot anticipate. It does not guarantee accuracy, remove the need for review, or make compute free.