Generative AI Leader Practice Exam 52 — 20 Questions

Question 1

A healthcare startup wants to use Vertex AI to deploy a model that helps doctors diagnose rare diseases. The model must be explainable, showing the reasoning path. Which technique should they implement?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Implement chain-of-thought prompting to output reasoning steps. Chain-of-thought reasoning allows the model to generate step-by-step explanations, which is crucial for medical diagnosis transparency.

Answer

Use a black-box ensemble model for higher accuracy

Answer

Reduce the model size to make it inherently interpretable

Answer

Disable all safety filters to avoid interfering with model output

Question 2

A financial services firm must comply with regulations when using gen AI. Which two measures are critical?

Accepted Answer

Implement audit trails. Audit trails are critical for compliance because they provide a tamper-evident, chronological record of all AI model inputs, outputs, and decisions. This enables firms to demonstrate regulatory adherence (e.g., under GDPR or SOX) by reconstructing the exact sequence of events that led to a specific AI-generated output, which is essential for accountability and forensic review.

Answer

Deploy without risk assessment

Answer

Use a closed-source model

Answer

Use only synthetic data

Question 3

A company is using Vertex AI to build a language model for generating legal documents. They need to ensure the model's outputs are accurate and verifiable. Which TWO features should they use?

Accepted Answer

Chain-of-thought reasoning. Chain-of-thought reasoning (C) is correct because it makes the model's step-by-step reasoning explicit, which improves the accuracy of complex legal drafting and lets reviewers trace how a conclusion or clause was derived, supporting verifiability. Grounding with citations to relevant legal texts (D) is correct because it anchors the model's outputs in authoritative sources and returns citations, so generated legal content can be checked against the referenced statutes, cases, or documents. The other options do not meet the requirement: confidence indicators (A) only give a score and do not make outputs verifiable; safety filters for legal content (B) restrict harmful or disallowed content but do not improve factual accuracy or traceability; and Model Cards (E) are documentation artifacts describing a model's intended use and limitations, not runtime features that verify generated outputs.

Answer

Confidence indicators

Answer

Safety filters for legal content

Answer

Model Cards

Question 4

A marketing agency uses a generative AI model to create slogans for ad campaigns. The model outputs generic slogans like 'Quality you can trust' that lack originality. The agency has a library of past award-winning slogans and wants to generate more creative and brand-specific outputs. They have a requirement that the model must not produce slogans longer than 15 words. Which technique should they prioritize?

Accepted Answer

Use few-shot prompting with 3-5 examples of award-winning slogans in the prompt.. Few-shot prompting (A) is the most direct and efficient technique because it provides the model with concrete examples of the desired output style (award-winning, creative slogans) within the context window, guiding the model's generation without altering its underlying weights. This approach immediately constrains the output to be brand-specific and creative by leveraging in-context learning, while the 15-word limit can be handled via a simple instruction in the prompt, avoiding the need for fine-tuning or risky parameter changes.

Answer

Set max tokens to 15 to force shorter, potentially more punchy slogans.

Answer

Increase the temperature to 1.2 to encourage more creative word combinations.

Answer

Fine-tune the model on the library of award-winning slogans.

Question 5

A healthcare organization wants to use generative AI for medical report summaries. What is the primary concern?

Accepted Answer

Ensuring HIPAA compliance and data security when using cloud AI services. The primary concern for a healthcare organization using generative AI for medical report summaries is ensuring HIPAA compliance and data security when using cloud AI services. Medical data is protected health information (PHI), and any cloud-based AI service must have a Business Associate Agreement (BAA) in place and enforce encryption at rest and in transit to avoid regulatory penalties and data breaches.

Answer

The model's ability to generate fluent and coherent summaries

Answer

Minimizing the cost of each API call to stay within budget

Answer

Latency of responses for real-time use cases

Question 6

A product team uses a translation model to convert English product descriptions into French. The model mixes formal and informal French dialects. Which simple prompt modification likely solves this?

Accepted Answer

Add a system prompt specifying 'Use only formal French with no informal expressions.'. Adding a system prompt that explicitly instructs the model to 'Use only formal French with no informal expressions' directly constrains the output style at inference time without requiring retraining. This leverages the model's instruction-following capability to enforce a specific dialect, which is the simplest and most effective modification for controlling output style in a production translation pipeline.

Answer

Increase the temperature to encourage more consistent output.

Answer

Fine-tune the model on a corpus of formal French texts.

Answer

Provide a few-shot example of a formal French translation in the prompt.

Question 7

A research team wants to generate high-quality images from text descriptions for a marketing campaign. They need the ability to edit specific regions of generated images while preserving the rest. Which Google Cloud AI service should they use?

Accepted Answer

Imagen on Vertex AI. Imagen on Vertex AI is Google Cloud's text-to-image generation service, and it supports advanced editing capabilities such as inpainting and outpainting, which allow specific regions of an image to be edited while preserving the rest. This directly matches the research team's requirement for high-quality image generation with region-specific editing.

Answer

Gemini 1.5 Pro with image input

Answer

Codey on Vertex AI

Answer

Veo on Vertex AI

Question 8

A financial services company wants to automate contract analysis to extract key clauses and identify risky terms. They have thousands of PDF contracts and need a solution that can be quickly integrated into their existing document management system. Which Google Cloud service is MOST suitable?

Accepted Answer

Document AI (DocAI). DocAI is purpose-built for document understanding and structured extraction from PDFs. Vertex AI Studio focuses on prompt design for generative models, while Model Garden is for selecting models. Vertex AI Agent Builder is for building conversational agents, not document processing.

Answer

Model Garden

Answer

Vertex AI Studio

Answer

Vertex AI Agent Builder

Question 9

A team is evaluating whether to use reinforcement learning from human feedback (RLHF) or in-context learning for a chatbot. Which TWO statements correctly describe trade-offs? (Select two.)

Accepted Answer

In-context learning is generally more cost-effective than RLHF for one-off tasks.. Option B is correct because in-context learning achieves task adaptation simply by placing examples or instructions in the prompt, requiring no gradient updates or preference dataset, so for a one-off task it avoids the substantial compute and labeling costs of RLHF. Option C is correct because RLHF typically involves collecting a large dataset of human preference comparisons, training a reward model on those preferences, and then fine-tuning the policy (e.g., via PPO) — an additional training pipeline beyond the base model. Option A is wrong because RLHF is defined by its use of human feedback; it cannot align behavior without human input. Option D is wrong because in-context learning leaves model weights unchanged, adapting behavior only through the prompt at inference time. Option E is wrong because context length is a property of the model architecture and serving setup, not a general advantage of in-context learning over RLHF fine-tuned models, which can share the same context window.

Answer

RLHF can align model behavior without any human input.

Answer

In-context learning permanently modifies model weights.

Answer

In-context learning can handle longer contexts than RLHF fine-tuned models.

Question 10

A research team uses a generative AI model to analyze historical texts. They want to provide users with insight into the model's reasoning process. Which explainability technique should they implement?

Accepted Answer

Chain-of-thought reasoning. Chain-of-thought reasoning provides step-by-step explanations of how the model arrived at its conclusion, enhancing transparency.

Answer

Grounding

Answer

Confidence indicators

Answer

Safety filters

Question 11

A financial services firm needs to use Gemini for analyzing customer transaction data. They require that all data remain within their VPC and that model inference logs be auditable. Which access tier should they choose?

Accepted Answer

Vertex AI. Vertex AI provides enterprise controls like VPC-SC, data isolation, and audit logging, while Google AI Studio is a prototyping environment without these guarantees.

Answer

Colab Enterprise

Answer

Gemini API without Vertex AI

Answer

Google AI Studio

Question 12

What is the primary function of embeddings in the context of generative AI?

Accepted Answer

To convert tokens into a fixed-length vector that captures semantic meaning. Embeddings are numerical representations of data that capture semantic meaning, enabling vector search and semantic similarity. They are not used for token generation directly, and they are not the same as temperature or context window.

Answer

To control the creativity of the model's output

Answer

To define the maximum length of the generated text

Answer

To determine the next token in a sequence during generation

Question 13

A team is deciding between using fine-tuning and in-context learning for a document classification task. They have 500 labeled examples and need low latency. Which TWO statements are accurate?

Accepted Answer

Fine-tuning generally improves accuracy more than in-context learning when sufficient labeled data is available. Option D is correct because fine-tuning updates the model's weights on the 500 labeled examples, allowing it to learn task-specific patterns and typically achieve higher accuracy than in-context learning when a sufficient amount of labeled data is available. Option E is correct because in-context learning places examples directly in the prompt at inference time, requiring no gradient updates or training step, so it can be used immediately with a pretrained model. Option A is incorrect because in-context learning increases prompt length and thus inference latency, so it does not always have lower latency than fine-tuning. Option B is incorrect because in-context learning works with any model context window and is not limited to windows smaller than 1000 tokens. Option C is incorrect because fine-tuning still requires a validation dataset to tune hyperparameters and monitor for overfitting.

Answer

In-context learning always has lower latency than fine-tuning

Answer

In-context learning can only be used with models that have a context window smaller than 1000 tokens

Answer

Fine-tuning eliminates the need for a validation dataset

Question 14

A company wants to use AI-generated images commercially. They are concerned about copyright and IP issues. Which action should they take FIRST to mitigate legal risk?

Accepted Answer

Ensure the AI model's training data does not contain copyrighted material without permission. The foundational legal risk in AI-generated imagery stems from the training data. If the model was trained on copyrighted works without permission, any output—even a novel image—can be considered a derivative work, exposing the company to infringement claims. Addressing the training data's compliance is the first and most critical step, as downstream mitigations (like watermarks or licenses) cannot retroactively fix an unlawfully trained model.

Answer

Conduct a freedom-to-operate search each time an image is generated

Answer

Add a watermark to all generated images using SynthID

Answer

Purchase a commercial license for the AI platform

Question 15

What is the fundamental difference between a foundation model and a fine-tuned model?

Accepted Answer

A foundation model is pre-trained on a large, diverse corpus; a fine-tuned model is adapted from a foundation model on a specific domain or task. A foundation model is pre-trained on broad data for general tasks, while a fine-tuned model is further trained on a specific dataset to specialize for a particular use case.

Answer

Foundation models only generate text; fine-tuned models can generate images

Answer

Foundation models are open-source; fine-tuned models are proprietary

Answer

Foundation models require no inference infrastructure; fine-tuned models do

Question 16

Which TWO options are best practices for deploying generative AI models on Vertex AI? (Choose two.)

Accepted Answer

Enable automatic scaling to handle variable traffic. Option B is correct because enabling automatic scaling on a Vertex AI Endpoint lets the deployed model add or remove replicas based on traffic and utilization, so it handles variable load efficiently while controlling cost during idle periods. Option C is correct because Vertex AI Model Monitoring continuously tracks prediction inputs and outputs against a baseline and alerts on training-serving skew and prediction drift, which is essential for maintaining generative AI model quality over time. Option A is wrong because disabling logging removes the audit trail and prediction telemetry needed for debugging, compliance, and monitoring, and the cost savings do not justify losing observability. Option D is wrong because manually scaling instances is error-prone and cannot react quickly to variable traffic the way autoscaling can. Option E is wrong because serving a model without optimization (for example, without appropriate machine types, batching, or quantization) wastes resources and degrades latency and throughput.

Answer

Disable logging to reduce cost

Answer

Manually scale instances based on expected load

Answer

Serve the model directly without optimization

Question 17

Which Google Cloud AI API is used to convert spoken language into text?

Accepted Answer

Speech-to-Text API. The Speech-to-Text API is Google Cloud's dedicated service for automatic speech recognition (ASR), converting audio input into written text using models like Chirp and the Universal speech model. It accepts audio from files or streaming sources and returns transcriptions with optional timestamps, speaker diarization, and word-level confidence. This directly matches the requirement of turning spoken language into text.

Answer

Text-to-Speech API

Answer

Natural Language API

Answer

Translation API

Question 18

An AI research lab wants to publish a model card for their new generative AI model. According to Google's Responsible AI practices, which information is ESSENTIAL to include in the model card?

Accepted Answer

The intended use, performance metrics, and known limitations. Model Cards should disclose intended use, performance metrics, and known limitations to promote transparency and accountability.

Answer

The exact architecture and hyperparameters used

Answer

A list of all training hardware and software versions

Answer

The names and salaries of the development team

Question 19

A company is fine-tuning a large language model for a domain-specific legal document summarization task. They have limited labeled data but want to adapt the model efficiently without catastrophic forgetting. Which technique is most suitable?

Accepted Answer

Supervised fine-tuning using LoRA adapters. LoRA (Low-Rank Adaptation) is the most suitable technique because it enables parameter-efficient fine-tuning by injecting trainable low-rank matrices into the transformer layers, drastically reducing the number of updated parameters. This preserves the pre-trained knowledge and prevents catastrophic forgetting, even with limited labeled data, while efficiently adapting the model to domain-specific legal summarization.

Answer

Reinforcement Learning from Human Feedback (RLHF)

Answer

In-context learning with few-shot examples in prompts

Answer

Full fine-tuning of all model parameters

Question 20

What is the primary purpose of the temperature parameter when using a generative language model?

Accepted Answer

It controls the randomness of token selection; lower values make output more deterministic. Temperature controls the randomness of token selection. Lower values produce more deterministic and conservative outputs, while higher values increase diversity and creativity.

Answer

It sets the maximum number of tokens the model can generate in a single response

Answer

It determines the number of candidate tokens considered at each step

Answer

It adjusts the similarity threshold for vector search retrieval