Courseiva

Google Cloud Generative AI Leader Generative AI Leader (Generative AI Leader) — Questions 226–300

683 questions total · 10pages · All types, answers revealed

Take a mock examExam hub
Page 3

Page 4 of 10

Page 5
226
MCQhard

A company is deploying a Gemini 1.0 Ultra model for a code generation assistant. They have set up Vertex AI Model Evaluation with a custom evaluation dataset to measure pass@1 accuracy. The initial evaluation shows 65% pass@1. They want to improve to 80% without collecting more training data. They have already attempted basic prompt engineering (e.g., 'write correct code') with limited improvement. Which approach is most likely to achieve the desired improvement?

A.Reduce the temperature to 0 and set top_p to 1.
B.Increase the number of output tokens and enable beam search with width 4.
C.Use chain-of-thought prompting with few-shot examples of correct code generation.
D.Apply reinforcement learning from human feedback (RLHF) using a reward model trained on the existing evaluation dataset.
AnswerC

Chain-of-thought elicits reasoning steps, improving accuracy beyond basic prompting.

Why this answer

Chain-of-thought prompting with few-shot examples is the most effective approach because it guides the model through step-by-step reasoning, which is critical for complex code generation tasks. This technique leverages the model's in-context learning ability to improve accuracy without additional training data, directly addressing the need to boost pass@1 from 65% to 80%.

Exam trap

Google Cloud often tests the misconception that hyperparameter tuning (like temperature or beam search) can substitute for structured prompting techniques, when in reality, chain-of-thought prompting directly addresses the reasoning gap that limits pass@1 accuracy in code generation.

How to eliminate wrong answers

Option A is wrong because reducing temperature to 0 and setting top_p to 1 makes the model deterministic, which may reduce diversity but does not inherently improve correctness for complex code generation; it can even cause repetitive or suboptimal outputs. Option B is wrong because increasing output tokens and enabling beam search with width 4 can improve exploration but does not guarantee higher pass@1 accuracy; beam search is more suited for tasks like translation and may not align with the goal of generating a single correct code snippet. Option D is wrong because applying RLHF requires a reward model trained on human preferences, not just the existing evaluation dataset, and this approach demands significant additional data and computational resources, contradicting the constraint of not collecting more training data.

Full explanation →
227
MCQmedium

A team is using a pre-trained language model to summarize legal documents. They find that summaries often miss key dates and parties involved. Which technique would most effectively improve factual accuracy?

A.Fine-tune the model on a dataset of legal summaries with annotated key entities.
B.Use top-p sampling with a low p value.
C.Increase the temperature parameter.
D.Use chain-of-thought prompting.
AnswerA

Fine-tuning adapts the model to domain-specific requirements, improving factual accuracy.

Why this answer

Fine-tuning on a dataset of legal summaries with annotated key entities directly teaches the model to recognize and reproduce critical factual elements like dates and parties. This supervised learning approach adjusts the model's weights to prioritize entity extraction and accurate generation, which is the most effective method for improving factual accuracy in domain-specific tasks.

Exam trap

Google Cloud often tests the misconception that inference-time parameters (temperature, top-p) or prompting strategies can substitute for targeted training, when in fact only fine-tuning with domain-specific annotated data reliably improves factual accuracy for structured entities.

How to eliminate wrong answers

Option B is wrong because top-p sampling with a low p value restricts the vocabulary to a small set of high-probability tokens, which can reduce creativity but does not address factual accuracy or entity recall—it may even omit rare but important entities. Option C is wrong because increasing the temperature parameter adds randomness to token selection, which typically reduces factual consistency and can lead to hallucinated or missing details. Option D is wrong because chain-of-thought prompting improves reasoning steps for multi-step tasks but does not inherently enforce factual accuracy for specific entities; it relies on the model's existing knowledge, which may still miss key dates and parties without targeted training.

Full explanation →
228
MCQhard

A company is piloting a GenAI code review assistant. Developers report that the assistant often suggests incorrect or insecure code snippets. The team wants to improve the assistant's reliability before expanding the pilot. Which approach should they prioritize?

A.Use RAG to retrieve code snippets from a database of known vulnerabilities
B.Increase the model's temperature parameter to generate more diverse suggestions
C.Switch to a larger foundation model without additional tuning
D.Fine-tune the base model on a curated dataset of secure code review examples
AnswerD

Fine-tuning on high-quality examples teaches the model to prioritize secure and correct suggestions.

Why this answer

Fine-tuning on a curated dataset of secure code review examples directly addresses the assistant's tendency to suggest incorrect or insecure code by adapting the model's behavior to the specific patterns and standards of secure coding. Unlike retrieval or parameter adjustments, this approach modifies the model's weights to prioritize security and correctness in its outputs, making it the most effective method for improving reliability in a targeted domain.

Exam trap

Google exams often test the misconception that adding more data (via RAG) or increasing model size (larger foundation model) automatically improves output quality, when in fact targeted fine-tuning on domain-specific, high-quality data is required to correct systematic errors in generative outputs.

How to eliminate wrong answers

Option A is wrong because RAG retrieves code snippets from a database of known vulnerabilities, which would surface insecure examples rather than correct ones, potentially worsening the assistant's suggestions. Option B is wrong because increasing the temperature parameter makes the model more random and less deterministic, which would increase the likelihood of generating diverse but even less reliable and more insecure code. Option C is wrong because switching to a larger foundation model without tuning does not guarantee improved security or correctness; larger models can still produce insecure code if not specifically aligned with secure coding practices, and they may introduce higher latency and cost without addressing the root cause.

Full explanation →
229
MCQeasy

A startup is developing a customer support chatbot using Vertex AI PaLM 2 API. They notice that the model sometimes generates plausible-sounding but factually incorrect information about company policies. The chatbot currently uses no external data. To reduce these hallucinations without retraining the model, the team needs a solution that can be implemented quickly and maintains low latency. They have access to the company's internal policy database stored in Cloud SQL. Which approach should they take?

A.Fine-tune the PaLM 2 model on a dataset of company policy documents.
B.Implement grounding by connecting the model to the company's policy database using Vertex AI Grounding.
C.Reduce the temperature parameter to 0 and increase top_k to 50.
D.Use prompt engineering to instruct the model to only answer from its internal knowledge.
AnswerB

Grounding directly ties responses to verified data, reducing hallucinations effectively.

Why this answer

Vertex AI Grounding connects the PaLM 2 model to the company's policy database in Cloud SQL, allowing the model to retrieve and cite factual information in real time. This approach reduces hallucinations without retraining, meets the low-latency requirement, and leverages existing internal data. Grounding works by augmenting the prompt with retrieved context from the grounding source, ensuring responses are factually grounded.

Exam trap

Google Cloud often tests the misconception that adjusting sampling parameters (temperature, top_k) can fix factual inaccuracies, when in reality those parameters only control creativity and randomness, not knowledge grounding.

How to eliminate wrong answers

Option A is wrong because fine-tuning requires a labeled dataset and retraining, which is time-consuming and does not meet the 'quickly implemented' requirement; it also does not guarantee real-time factual accuracy for dynamic policies. Option C is wrong because reducing temperature to 0 and increasing top_k to 50 only affects output randomness and diversity, not factual grounding—hallucinations stem from lack of external knowledge, not sampling parameters. Option D is wrong because prompt engineering alone cannot prevent the model from generating plausible-sounding falsehoods; the model's internal knowledge is static and may be outdated or incorrect, and it cannot access the Cloud SQL database without a retrieval mechanism.

Full explanation →
230
MCQmedium

An organization uses an IAM policy for Vertex AI as shown. A security audit reveals that engineer@example.com deployed a model that inadvertently exposed sensitive data. What is the most likely reason this happened?

A.Audit logging is not enabled for DATA_WRITE events.
B.The admin user did not review the deployment.
C.The engineer had the aiplatform.user role, which includes permissions to deploy models without additional review.
D.The policy does not include a separation of duties between development and production.
AnswerC

The user role allows deployment, and no approval gate is enforced.

Why this answer

The `aiplatform.user` role in Vertex AI includes the `aiplatform.model.deploy` permission, which allows any user with that role to deploy models without requiring additional approvals or administrative review. This lack of a secondary authorization step means the engineer could deploy a model that exposed sensitive data, even if the model had not been properly vetted for data leakage.

Exam trap

The trap here is that candidates may focus on operational failures like missing audit logs or lack of review, rather than recognizing that the IAM role itself grants the permission to deploy without any guardrails, which is the direct technical cause of the exposure.

How to eliminate wrong answers

Option A is wrong because audit logging for DATA_WRITE events records actions after they occur, but does not prevent the deployment itself; the exposure happened due to insufficient permissions control, not missing logs. Option B is wrong because the admin user not reviewing the deployment is a process failure, but the root cause is that the IAM policy granted the engineer the ability to deploy without any review being required. Option D is wrong because while separation of duties is a best practice, the specific IAM policy shown does not enforce it; the question asks for the most likely reason the exposure occurred, which is the direct permission granted by the `aiplatform.user` role.

Full explanation →
231
MCQmedium

A team is building a generative AI model for customer support. They notice the model often produces overly polite but unhelpful responses. Which technique would best improve response quality without sacrificing helpfulness?

A.Apply reinforcement learning from human feedback (RLHF)
B.Increase the amount of training data
C.Lower the top_k sampling value
D.Increase the temperature parameter
AnswerA

RLHF tunes the model to align with desired response characteristics.

Why this answer

RLHF directly addresses the misalignment between the model's training objective (e.g., predicting the next token) and the desired outcome (helpful, not just polite). By using human feedback to train a reward model, the system learns to optimize for response quality and helpfulness, reducing sycophantic or overly polite but uninformative outputs.

Exam trap

Google Cloud often tests the misconception that hyperparameter tuning (temperature, top_k) or more data alone can fix alignment issues, when in fact only RLHF directly optimizes for human-judged helpfulness and quality.

How to eliminate wrong answers

Option B is wrong because simply increasing training data does not correct the model's tendency toward polite but unhelpful responses; it may reinforce existing patterns without addressing alignment. Option C is wrong because lowering top_k sampling reduces diversity by restricting token choices to the top k most likely tokens, which can make responses even more generic and less helpful, not more substantive. Option D is wrong because increasing the temperature parameter increases randomness in token selection, which can lead to less coherent or more erratic responses, not more helpful ones.

Full explanation →
232
MCQmedium

A company uses Gemini for Google Workspace to help employees draft emails. After rollout, the IT team notices that the Smart Compose suggestions are sometimes inappropriate for external communications. What is the BEST way to address this while maintaining productivity gains?

A.Switch to a third-party email assistant not integrated with Workspace
B.Train employees to write more specific prompts and provide few-shot examples for desired tone
C.Disable Smart Compose for all external email accounts
D.Fine-tune the Gemini model on the company's past external emails
AnswerB

Prompt engineering training helps employees guide the model effectively, reducing inappropriate suggestions while maintaining productivity.

Why this answer

Providing few-shot examples or using custom prompts for tone can guide the model to produce more appropriate suggestions. Disabling the feature or using a generic model would not leverage Workspace integration.

Full explanation →
233
MCQhard

A company is using a fine-tuned LLM for generating technical support responses. After deployment, they notice that the model sometimes produces incorrect but plausible-sounding answers (hallucinations). They have a large repository of verified technical manuals. Which technique would BEST reduce hallucinations while minimizing the need for additional training?

A.Increase the temperature parameter to make the model more conservative
B.Use a larger base model with a longer context window
C.Fine-tune the model again with a larger dataset of verified responses
D.Implement RAG by indexing the verified manuals and retrieving relevant sections during inference
AnswerD

RAG provides the model with factual context from the manuals at inference time, significantly reducing hallucinations without retraining.

Why this answer

RAG with the verified manuals as the knowledge base allows the model to ground its responses in authoritative sources without retraining. Fine-tuning might still hallucinate if the data is insufficient. Prompt engineering alone cannot eliminate hallucinations.

Using a larger context window may help but is not as reliable as RAG.

Full explanation →
234
Multi-Selecteasy

Which THREE components are core to a typical Retrieval Augmented Generation (RAG) system?

Select 3 answers
A.Classifier
B.Vector database
C.Embedding model
D.Rewriter
E.Large language model
AnswersB, C, E

Stores embeddings and enables similarity search.

Why this answer

A vector database (B) is core to a RAG system because it stores and indexes embeddings of external knowledge chunks, enabling efficient similarity search to retrieve the most relevant context for a user query. This retrieved context is then provided to the LLM to ground its response in factual data, reducing hallucinations and improving accuracy.

Exam trap

Google Cloud often tests the distinction between core mandatory components (embedding model, vector DB, LLM) and optional auxiliary components (classifier, rewriter, reranker) to see if candidates understand the minimal viable RAG architecture versus extended pipelines.

Full explanation →
235
Multi-Selecteasy

Which TWO techniques are commonly used to control the style and tone of a generative model's output?

Select 2 answers
A.Adjusting the temperature
B.Modifying the top_k value
C.Fine-tuning on a dataset with desired style
D.Prompt engineering with style instructions
E.Changing the top_p value
AnswersC, D

Fine-tuning adapts the model to a specific style.

Why this answer

Fine-tuning on a dataset that embodies the desired style directly adjusts the model's weights, making it consistently produce outputs with that specific tone and style. This is a fundamental technique for customizing generative models, as it teaches the model the exact patterns, vocabulary, and stylistic nuances present in the training data.

Exam trap

Google Cloud often tests the distinction between sampling parameters (temperature, top_k, top_p) that control output randomness and diversity versus training or conditioning techniques (fine-tuning, prompt engineering) that directly influence style and tone, leading candidates to incorrectly select sampling parameters as style-control methods.

Full explanation →
236
MCQeasy

A sales team wants to use Gemini for Google Workspace to draft emails in Gmail. Which feature should they enable to generate full email drafts from a brief prompt?

A.Translate
B.Smart Reply
C.Help me write
D.Smart Compose
AnswerC

Help me write generates a complete email draft from a prompt, ideal for drafting.

Why this answer

'Help me write' is the specific Gemini for Google Workspace feature designed to generate full email drafts from a brief prompt. It leverages the underlying large language model to compose complete, context-aware email content, unlike features that only suggest short phrases or replies.

Exam trap

The trap here is that candidates often confuse 'Smart Compose' (which only suggests short text completions) with the full-draft generation capability of 'Help me write', leading them to select the wrong feature.

How to eliminate wrong answers

Option A is wrong because 'Translate' is a language translation feature, not a generative drafting tool; it converts existing text from one language to another. Option B is wrong because 'Smart Reply' only suggests short, pre-generated response phrases (e.g., 'Yes, sounds good') based on the email content, not full drafts from a prompt. Option D is wrong because 'Smart Compose' provides inline word and phrase suggestions as you type, but does not generate a complete email draft from a brief prompt.

Full explanation →
237
MCQeasy

According to Google's AI Principles, AI systems should be 'accountable to people.' What does this principle primarily require?

A.AI systems should always have a human monitor present during operation
B.The AI system should be able to explain all its decisions to end users
C.Organizations should provide avenues for feedback and recourse when AI systems cause harm
D.AI systems should be designed to maximize shareholder value
AnswerC

This directly aligns with the accountability principle.

Why this answer

Accountability means providing mechanisms for feedback and recourse when AI systems cause harm or make mistakes.

Full explanation →
238
MCQeasy

A small marketing agency with 10 employees is exploring generative AI to create personalized ad copy for their clients. They have a limited budget of $5,000 per month and no in-house machine learning expertise. The CEO wants to have a working prototype within two weeks to show to a potential client. The agency's data is sensitive and cannot be shared with unauthorized third parties. Which strategy should they pursue?

A.Hire a team of data scientists to fine-tune an open-source model
B.Use a third-party platform that requires on-premise deployment
C.Build a custom foundation model from scratch using their client data
D.Use Google's Generative AI Studio with pre-trained models via API
AnswerD

Managed service enables quick, low-cost prototyping with data privacy.

Why this answer

Google's Generative AI Studio provides pre-trained models via API, allowing the agency to quickly prototype personalized ad copy without needing in-house ML expertise. This approach respects the $5,000 budget (API usage is cost-effective for small-scale prototyping), meets the two-week timeline (no training required), and ensures data privacy by using Google Cloud's data governance controls (data is not shared with unauthorized third parties).

Exam trap

Google Cloud often tests the misconception that building or fine-tuning a model from scratch is the only way to achieve customization, when in fact pre-trained APIs with prompt engineering or lightweight fine-tuning can meet business constraints like budget, timeline, and expertise.

How to eliminate wrong answers

Option A is wrong because hiring a team of data scientists to fine-tune an open-source model would exceed the $5,000 monthly budget and the two-week timeline, and the agency lacks the in-house expertise to manage such a team. Option B is wrong because requiring on-premise deployment contradicts the agency's lack of ML expertise and limited budget; on-premise solutions typically involve high upfront costs and ongoing maintenance. Option C is wrong because building a custom foundation model from scratch is prohibitively expensive (often millions of dollars), requires vast amounts of data and compute resources, and cannot be completed within two weeks or within a $5,000 budget.

Full explanation →
239
MCQhard

A company is deploying a generative AI application that must comply with GDPR's right to explanation. The application must be able to justify its decisions. Which model or approach provides the MOST inherent interpretability?

A.Use Gemini 1.5 Pro with a system prompt asking for explanations
B.Use a smaller, simpler model that is inherently more interpretable, such as a logistic regression or decision tree
C.Use Retrieval-Augmented Generation (RAG) to ground responses in source documents
D.Use PaLM 2 with prompt engineering to provide step-by-step reasoning
AnswerB

Simpler models are transparent and decisions can be directly traced, satisfying the right to explanation more reliably.

Why this answer

Smaller, simpler models are inherently more interpretable. Large black-box models (Gemini, PaLM) are difficult to explain. RAG improves factual grounding but not interpretability of the model's reasoning.

Prompting does not make the model's internal logic transparent.

Full explanation →
240
MCQmedium

A security team wants to prevent prompt injection attacks on their generative AI application hosted on Vertex AI. Which best practice should they implement?

A.Use a custom model instead of a foundation model
B.Disable all logging
C.Use a private endpoint
D.Implement input validation and output filtering
AnswerD

This helps detect and block malicious prompts and undesired outputs.

Why this answer

Prompt injection attacks exploit the model's inability to distinguish between user instructions and untrusted input. Implementing input validation (e.g., sanitizing special characters or known injection patterns) and output filtering (e.g., using a classifier to detect and block malicious responses) directly mitigates this risk by controlling what the model processes and returns. On Vertex AI, this can be enforced via custom safety attributes or integration with services like Cloud DLP for data loss prevention.

Exam trap

The trap here is that candidates confuse network-level security controls (like private endpoints) with application-layer security controls, assuming that restricting network access alone can prevent content-based attacks like prompt injection.

How to eliminate wrong answers

Option A is wrong because using a custom model does not inherently prevent prompt injection; the vulnerability exists in any model that processes untrusted input, regardless of whether it is a foundation model or a custom model. Option B is wrong because disabling logging removes visibility into attack attempts and compliance auditing, but does not prevent the injection itself; logging is a detection mechanism, not a prevention control. Option C is wrong because a private endpoint (e.g., Private Service Connect) secures network traffic by keeping it within a VPC, but it does not inspect or sanitize the content of prompts or outputs, leaving the application vulnerable to injection attacks at the application layer.

Full explanation →
241
MCQeasy

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

A.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
B.Foundation models only generate text; fine-tuned models can generate images
C.Foundation models are open-source; fine-tuned models are proprietary
D.Foundation models require no inference infrastructure; fine-tuned models do
AnswerA

Foundation models like Gemini are pre-trained broadly; fine-tuning adapts them for specialized tasks.

Why this answer

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.

Full explanation →
242
MCQeasy

A company wants to deploy a chatbot for customer support. To comply with the AI Principle of being 'accountable to people', which practice should they implement?

A.Provide human review for high-stakes decisions made by the chatbot
B.Allow the chatbot to escalate all conversations to human agents
C.Ensure the chatbot can override human agents' decisions
D.Disable logging of chatbot interactions to protect user privacy
AnswerA

Human review upholds accountability by allowing people to oversee and intervene when necessary.

Why this answer

Providing human oversight for high-stakes decisions ensures accountability, aligning with Google's AI Principle that AI systems should be accountable to people.

Full explanation →
243
MCQhard

A data scientist is using Vertex AI to fine‑tune a PaLM 2 model for a legal document summarization task. They have 10,000 labeled document‑summary pairs. After supervised fine‑tuning, the model performs well on the training set but often hallucinates names and dates on unseen documents. Which next step is MOST likely to improve factual accuracy?

A.Increase the number of fine‑tuning epochs to 10
B.Use a larger base model like Gemini Ultra without fine‑tuning
C.Apply reinforcement learning from human feedback (RLHF) using a preference dataset that penalizes factual inaccuracies
D.Add more training examples from a different domain
AnswerC

RLHF can directly optimize for factual correctness by rewarding accurate summaries and penalizing hallucinations.

Why this answer

RLHF (Reinforcement Learning from Human Feedback) is specifically designed to align model outputs with human preferences, which can reduce hallucinations by penalizing factually incorrect generations. More data or longer training may not fix the underlying alignment issue. RAG is a separate approach but RLHF directly addresses the hallucination from the model's behavior.

Full explanation →
244
Multi-Selecthard

A company deploys a Gemini model on Vertex AI for a customer-facing chatbot. They observe the chatbot occasionally produces toxic language. Which TWO measures should they implement immediately to reduce toxic outputs?

Select 2 answers
A.Increase the model's temperature to make outputs more conservative.
B.Use a separate language model to rephrase the outputs before sending to users.
C.Fine-tune the model on a curated dataset of polite conversations.
D.Enable the 'block offensive content' flag in the model's safety configuration.
E.Configure the safety thresholds in the Vertex AI endpoint deployment to block hate speech and toxic content.
AnswersD, E

This flag directly enables content filtering.

Why this answer

Enabling the 'block offensive content' flag directly activates Gemini's built-in safety filters, which are designed to detect and suppress toxic language at the model's output layer. This is an immediate, configuration-level measure that requires no additional training or external services, making it the fastest way to reduce harmful responses in a production chatbot.

Exam trap

Google Cloud often tests the misconception that increasing temperature or fine-tuning are quick fixes for safety issues, when in fact they are either counterproductive or require significant time and resources, whereas safety configuration flags are the immediate, recommended first step.

Full explanation →
245
MCQmedium

A company wants to ground Gemini responses with real-time Google Search results to improve accuracy of current events. Which feature enables this?

A.BigQuery ML
B.Vertex AI Vector Search
C.Gemini API function calling
D.Google Search grounding
AnswerD

This feature links Gemini to Google Search.

Why this answer

Google Search grounding connects Gemini to live search results to reduce hallucinations and provide up-to-date information.

Full explanation →
246
Multi-Selectmedium

Which TWO factors are most critical when deciding to build a custom GenAI model vs. using a pre-built API? (Select two.)

Select 2 answers
A.Availability of in-house ML talent
B.Need for domain-specific knowledge
C.Number of layers in the model
D.Brand reputation of the model provider
E.Volume of expected inference requests
AnswersA, B

Building a custom model requires significant ML expertise; without it, using an API is more practical.

Why this answer

Building a custom GenAI model requires specialized machine learning expertise, including proficiency in frameworks like PyTorch or TensorFlow, experience with distributed training (e.g., using Horovod or DeepSpeed), and the ability to fine-tune architectures like transformers. Without in-house ML talent, the organization cannot effectively manage data curation, hyperparameter tuning, or model evaluation, making a pre-built API the more viable choice. This factor directly determines whether the organization has the technical capacity to undertake custom development.

Exam trap

Google Cloud often tests the distinction between strategic business factors (like in-house talent and domain specificity) versus operational or vendor-related details (like model layers, brand reputation, or request volume) to see if candidates can separate high-level decision drivers from low-level implementation concerns.

Full explanation →
247
MCQhard

A data scientist is fine-tuning a large language model for a legal document summarization task. The dataset contains only 500 examples, and the model must not forget its general language capabilities. Which fine-tuning method is most suitable?

A.Retraining the model from scratch on the legal dataset
B.Adapter-based fine-tuning using LoRA
C.In-context learning with a few examples in the prompt
D.Full fine-tuning of all model parameters
AnswerB

LoRA fine-tunes a small set of adapter parameters, preserving general knowledge while adapting to the specific task with a small dataset.

Why this answer

LoRA (Low-Rank Adaptation) is an adapter-based fine-tuning method that updates a small number of parameters while keeping the base model frozen. This prevents catastrophic forgetting and works well with small datasets. Full fine-tuning would risk overfitting and losing general capabilities.

Full explanation →
248
MCQeasy

An organization wants to ensure their generative AI application does not produce toxic or harmful content. Which Vertex AI feature should they implement?

A.Safety filters and content moderation
B.Explainable AI
C.AutoML
D.Model Monitoring
AnswerA

These features are designed to detect and mitigate harmful content in model outputs.

Why this answer

Safety filters and content moderation in Vertex AI allow organizations to define and enforce policies that block or flag toxic, harmful, or inappropriate content generated by the model. This feature uses pre-built and customizable classifiers to evaluate prompts and responses against safety attributes (e.g., hate speech, harassment, sexually explicit content) before returning them to the user, directly addressing the requirement to prevent harmful outputs.

Exam trap

Google Cloud often tests the distinction between features that *analyze* model behavior (like Explainable AI or Model Monitoring) versus features that *actively enforce* safety policies (like Safety Filters), leading candidates to confuse monitoring or interpretability tools with content moderation controls.

How to eliminate wrong answers

Option B (Explainable AI) is wrong because it focuses on interpreting model predictions (e.g., feature attributions) rather than blocking toxic content; it provides transparency but no active content filtering. Option C (AutoML) is wrong because it automates model training and deployment for custom ML tasks, not content moderation or safety enforcement. Option D (Model Monitoring) is wrong because it tracks model performance and drift over time (e.g., prediction skew, data drift), not real-time content safety checks on individual outputs.

Full explanation →
249
MCQmedium

A company wants to build a customer service chatbot that answers questions about their internal policy documents. The documents are updated monthly, and the team cannot afford to retrain a model each time. Which approach is MOST appropriate?

A.Use a larger foundation model with a longer context window and paste all documents into each prompt
B.Fine-tune a base LLM on the policy documents monthly
C.Train a custom model from scratch on the policy documents each month
D.Use Retrieval-Augmented Generation (RAG) with the policy documents indexed in a vector store
AnswerD

RAG retrieves relevant document chunks at query time, ensuring the chatbot always answers from the latest uploaded documents without any model retraining.

Why this answer

Retrieval-Augmented Generation (RAG) is the most appropriate approach because it allows the chatbot to answer questions based on the latest policy documents without retraining the model. By indexing the documents in a vector store and retrieving relevant chunks at inference time, RAG ensures the model's responses are grounded in the current data, accommodating monthly updates efficiently.

Exam trap

Google often tests the misconception that fine-tuning is the only way to incorporate new data, when in fact RAG provides a dynamic, cost-effective alternative that avoids retraining.

How to eliminate wrong answers

Option A is wrong because pasting all documents into each prompt would exceed the context window limits of even the largest foundation models (e.g., 128K tokens), leading to high latency, cost, and potential truncation of critical information. Option B is wrong because fine-tuning a base LLM monthly on the policy documents is computationally expensive, time-consuming, and risks catastrophic forgetting of previously learned knowledge, making it unsustainable for frequent updates. Option C is wrong because training a custom model from scratch each month is prohibitively expensive, requires massive datasets and compute resources, and is entirely unnecessary when RAG can achieve the same goal with far less overhead.

Full explanation →
250
MCQeasy

A company wants to ensure that AI-generated images can be traced back to their origin to prevent misuse. Which Google technology provides invisible watermarking for AI-generated content?

A.Datasheets for Datasets
B.SynthID
C.Model Cards
D.People + AI Guidebook
AnswerB

SynthID is specifically designed to watermark AI-generated content imperceptibly for traceability.

Why this answer

SynthID is Google DeepMind's technology for embedding invisible watermarks into AI-generated images, audio, text, and video.

Full explanation →
251
Multi-Selectmedium

A fintech company is deploying a generative AI system that offers investment advice. To comply with regulations and Google's AI Principles, they need to ensure appropriate human oversight and transparency. Which two actions should they take? (Choose two.)

Select 2 answers
A.Use only synthetic data for training to avoid privacy issues
B.Allow the AI to execute trades automatically without human intervention
C.Require a human advisor to review and approve all AI-generated recommendations
D.Develop and publish a Model Card that describes the model's limitations and intended use
E.Disable logging of all interactions to maximize user privacy
AnswersC, D

Why this answer

Human review for high-stakes decisions (C) ensures accountability, and providing model cards (D) ensures transparency about capabilities and limitations. Publishing a Model Card is a key practice to comply with Google's AI Principles and regulatory requirements for transparency.

Full explanation →
252
MCQhard

A company wants to use generative AI for creative content generation (e.g., marketing copy). They need to ensure the content is original and does not plagiarize existing materials. Which combination of strategies is most effective?

A.Use a model with a high temperature setting and post-process with plagiarism checker.
B.Fine-tune the model on a dataset of already-created content to learn style.
C.Use a retrieval-augmented generation system that explicitly avoids copying.
D.Limit the model to generate only short snippets.
AnswerC

RAG can be configured to paraphrase or generate novel content while staying relevant, reducing plagiarism risk.

Why this answer

Retrieval-augmented generation (RAG) systems explicitly retrieve relevant, non-copyrighted or licensed content from a curated knowledge base and generate outputs grounded in that retrieved data, which inherently reduces the risk of verbatim copying. Unlike simple plagiarism checkers or temperature adjustments, RAG combines retrieval with generation to ensure originality by design, making it the most effective strategy for avoiding plagiarism in creative content generation.

Exam trap

Google Cloud often tests the misconception that randomness (high temperature) or post-processing (plagiarism checkers) can prevent plagiarism, when in fact only retrieval-augmented generation or similar grounding techniques address the root cause of copying from training data.

How to eliminate wrong answers

Option A is wrong because high temperature settings increase randomness and creativity but do not prevent the model from memorizing and reproducing training data verbatim; a post-process plagiarism checker can only detect copying after generation, not prevent it, and may miss paraphrased or structurally similar content. Option B is wrong because fine-tuning on already-created content teaches the model to mimic existing styles and patterns, which increases the risk of overfitting and reproducing copyrighted or plagiarized material, especially if the dataset contains protected works. Option D is wrong because limiting output length does not address the core issue of originality; short snippets can still be direct copies of existing phrases or sentences, and the strategy fails to ensure content is novel or properly attributed.

Full explanation →
253
MCQeasy

A retail company wants to build a customer service chatbot that can handle returns, order status, and FAQs. They need to integrate with their existing backend systems. Which Google Cloud service should they use?

A.Vertex AI Model Garden
B.Vertex AI Agent Builder
C.Vertex AI Search
D.Vertex AI Codey API
AnswerB

Provides tools for building chatbots with backend integration.

Why this answer

Vertex AI Agent Builder is the correct choice because it provides a low-code platform specifically designed for building conversational AI agents (chatbots) that can be integrated with enterprise backend systems via APIs, connectors, and custom tools. It supports grounding in enterprise data, multi-turn dialogue management, and seamless integration with existing systems for handling returns, order status, and FAQs, making it the most suitable service for this use case.

Exam trap

The trap here is that candidates may confuse Vertex AI Agent Builder with Vertex AI Search or Model Garden, assuming any generative AI service can build a chatbot, but only Agent Builder provides the necessary conversational orchestration and backend integration capabilities required for a production customer service chatbot.

How to eliminate wrong answers

Option A is wrong because Vertex AI Model Garden is a repository of pre-trained and foundation models for discovery and deployment, not a service for building conversational agents with backend integration. Option C is wrong because Vertex AI Search is optimized for enterprise search and information retrieval over structured and unstructured data, not for building multi-turn conversational chatbots that require backend system integration. Option D is wrong because Vertex AI Codey API is focused on code generation and code-related tasks (e.g., code completion, chat, and generation), not on building customer service chatbots that interact with backend systems.

Full explanation →
254
MCQmedium

A data analyst wants to experiment with ML models using a free, cloud-based Jupyter notebook environment with GPU support. Which Google tool should they use?

A.Cloud Shell
B.Google Colab
C.BigQuery Studio
D.Vertex AI Workbench
AnswerB

Google Colab offers free notebooks with GPU support for ML experimentation.

Why this answer

Google Colab is a free, cloud-based Jupyter notebook environment that provides access to GPUs (e.g., NVIDIA T4 or V100) without requiring any setup or billing. It is specifically designed for experimentation and learning with ML models, making it the correct choice for a data analyst seeking a free, GPU-enabled notebook environment.

Exam trap

The trap here is that candidates may confuse Vertex AI Workbench (a paid, enterprise tool) with a free offering because both provide Jupyter notebooks with GPU support, but the question's requirement for a 'free' tool eliminates it.

How to eliminate wrong answers

Option A is wrong because Cloud Shell is a browser-based terminal with a small, ephemeral VM (typically 5 GB of persistent disk) that does not include GPU support or a Jupyter notebook interface; it is meant for command-line operations and not for running ML experiments. Option C is wrong because BigQuery Studio is a data analysis and SQL workspace within BigQuery that focuses on querying large datasets, not on running ML model training with GPU acceleration in a notebook environment. Option D is wrong because Vertex AI Workbench is a fully managed, enterprise-grade Jupyter notebook environment that does support GPUs, but it requires a paid Google Cloud project and is not free; the question explicitly asks for a free tool.

Full explanation →
255
MCQeasy

What is the primary advantage of using embeddings and vector search for semantic search over traditional keyword search?

A.Faster retrieval speed
B.Lower storage requirements
C.No need for indexing
D.Ability to find documents with similar meaning even without exact keyword matches
AnswerD

Semantic search using embeddings captures context and synonyms.

Why this answer

Embeddings and vector search capture semantic meaning by converting text into high-dimensional vectors, enabling retrieval of documents with similar meaning even when they lack exact keyword matches. This is the primary advantage over traditional keyword search, which relies on literal term matching and fails with synonyms, paraphrases, or conceptual similarity.

Exam trap

Google often tests the misconception that vector search is faster or requires less storage than keyword search, but the real advantage is semantic understanding, not performance or resource efficiency.

How to eliminate wrong answers

Option A is wrong because vector search is typically slower than keyword search due to the computational cost of approximate nearest neighbor (ANN) algorithms and high-dimensional distance calculations, whereas keyword search uses inverted indexes for near-instant term lookups. Option B is wrong because embeddings and vector indexes often require more storage than keyword indexes, as each document is represented by a dense vector (e.g., 768 or 1536 dimensions) plus the index structure, while keyword indexes store sparse term-document mappings. Option C is wrong because vector search still requires indexing—specifically, building an ANN index (e.g., HNSW, IVF) to enable efficient similarity search; without indexing, a brute-force scan of all vectors would be impractical at scale.

Full explanation →
256
Multi-Selecteasy

Which TWO are benefits of using pre-trained foundation models instead of training from scratch?

Select 2 answers
A.Complete control over model architecture
B.Lower training cost
C.Eliminates the need for prompt engineering
D.Guaranteed absence of bias
E.Faster time to deployment
AnswersB, E

Pre-trained models require less compute and data, reducing cost.

Why this answer

Pre-trained foundation models have already been trained on vast datasets, so you only need to fine-tune them for your specific task. This dramatically reduces the compute resources, time, and data required compared to training from scratch, directly lowering training cost.

Exam trap

Google Cloud often tests the misconception that pre-trained models eliminate the need for any further engineering (like prompt engineering) or that they are completely bias-free, when in fact they still require careful tuning and can perpetuate biases from their training data.

Full explanation →
257
MCQeasy

Refer to the exhibit. A user wants formal translations from a generative AI model, but the model outputs informal style inconsistently. Which prompt engineering technique would best ensure consistent formal translations?

A.Use context caching
B.Provide a few-shot example with formal and informal pairs
C.Use a longer system prompt with detailed rules
D.Set top_k to 1
AnswerB

Correct: Few-shot examples directly show the expected output format.

Why this answer

Providing a few-shot example that explicitly demonstrates the desired formal translation guides the model to follow that pattern. System instructions can help but are less direct.

Full explanation →
258
MCQmedium

A company uses a text-to-image model to generate marketing visuals. The results often misinterpret the prompt, e.g., 'a red car' generates a blue car. Which technique should they try first to align the output with the prompt?

A.Use a negative prompt to exclude blue
B.Refine the prompt with more adjectives and context, e.g., 'bright red sports car'
C.Upscale the image resolution to 1024x1024
D.Increase the guidance scale to 20
AnswerB

Clearer, more descriptive prompts help the model understand the desired output.

Why this answer

Refining the prompt with more adjectives and context directly addresses the root cause of misalignment: insufficient specificity in the text description. Text-to-image models rely on the semantic richness of the prompt to guide the latent diffusion process; adding 'bright red sports car' provides stronger conditioning signals that steer the model's cross-attention layers toward the intended color and object attributes. This is the most efficient first step before adjusting hyperparameters like guidance scale.

Exam trap

The trap here is that candidates often jump to hyperparameter tuning (guidance scale) or post-processing (upscaling) as a first fix, when the most fundamental and cost-effective step is to improve the input prompt's specificity, which directly controls the conditioning signal in the diffusion process.

How to eliminate wrong answers

Option A is wrong because using a negative prompt to exclude 'blue' is a reactive band-aid that does not fix the core issue of the model failing to associate 'red' with the car; it also risks suppressing other unintended features and can degrade image quality by over-constraining the latent space. Option C is wrong because upscaling resolution to 1024x1024 only increases pixel density and does not alter the semantic alignment between the prompt and the generated image; the model's misinterpretation of 'red' would persist at any resolution. Option D is wrong because increasing the guidance scale to 20 excessively amplifies the prompt's influence, often leading to image saturation, artifacts, and mode collapse, while still not correcting the fundamental misassociation of the color attribute.

Full explanation →
259
MCQhard

A global company deploys a generative AI chatbot in the European Union. They must ensure compliance with GDPR regarding user data used for fine-tuning the model. What is the MOST important requirement they must fulfill?

A.Use only synthetic data for fine-tuning to avoid GDPR issues
B.Store all fine-tuned models only on US-based servers
C.Anonymize all data before fine-tuning, regardless of consent
D.Obtain explicit consent from data subjects for using their data in fine-tuning
AnswerD

Explicit consent is a lawful basis under GDPR for processing personal data, especially for secondary uses like fine-tuning.

Why this answer

Under GDPR, using personal data for fine-tuning a generative AI model constitutes a new processing purpose that requires a lawful basis. Explicit consent (Article 7 and Article 9) is the most robust basis when relying on consent, as it must be freely given, specific, informed, and unambiguous. Without explicit consent, the company risks violating data minimization and purpose limitation principles, even if other anonymization or storage measures are applied.

Exam trap

The Generative AI Leader exam often tests the misconception that technical measures like anonymization or data localization can substitute for a proper lawful basis under GDPR, when in fact the lawful basis (such as explicit consent) is the foundational requirement that must be established before any processing begins.

How to eliminate wrong answers

Option A is wrong because synthetic data can still be derived from real personal data and may retain patterns that re-identify individuals; GDPR does not exempt synthetic data from compliance if it is generated from personal data. Option B is wrong because storing models on US-based servers does not address GDPR requirements—data transfer must comply with adequacy decisions or Standard Contractual Clauses (SCCs), and server location alone does not provide a lawful basis for processing. Option C is wrong because anonymization is a valid approach only if it is irreversible and meets the GDPR's anonymization standard (Recital 26); however, it does not replace the need for a lawful basis if the data was originally collected for a different purpose, and consent may still be required for the fine-tuning activity.

Full explanation →
260
MCQhard

A team is building a multilingual customer support chatbot using Gemini. They notice that for low-resource languages, the model frequently produces grammatically incorrect responses. Which strategy would MOST effectively improve quality for these languages without sacrificing latency?

A.Implement a separate translation step: translate user input to English, generate response in English, then translate back
B.Fine-tune the model on a large parallel corpus for each low-resource language
C.Use few-shot prompting with high-quality examples in the target language
D.Increase temperature to 1.5 to encourage more diverse grammar
AnswerC

Few-shot examples can improve output by demonstrating correct grammar and style, and do not add latency beyond the prompt length.

Why this answer

Few-shot prompting with high-quality examples in the target language directly conditions the model to produce grammatically correct responses for low-resource languages without adding inference latency. This approach leverages Gemini's in-context learning capability, allowing it to adapt to the target language's grammar patterns without requiring additional model training or translation round-trips.

Exam trap

This question tests the misconception that translation pipelines or fine-tuning are the only ways to handle low-resource languages, but the trap here is that few-shot prompting can achieve comparable quality improvements with zero additional latency and minimal data requirements.

How to eliminate wrong answers

Option A is wrong because implementing a separate translation step (translate to English, generate, translate back) introduces significant latency from two additional inference calls and risks compounding translation errors, especially for low-resource languages where translation quality is poor. Option B is wrong because fine-tuning on a large parallel corpus for each low-resource language is resource-intensive, requires substantial labeled data that may not exist, and does not guarantee latency preservation since fine-tuning does not affect inference speed but the data requirement is impractical. Option D is wrong because increasing temperature to 1.5 encourages more diverse but often less coherent outputs, which would exacerbate grammatical errors in low-resource languages rather than improve them.

Full explanation →
261
MCQeasy

A developer wants to use Gemini 1.5 Pro to analyze hour-long video content and generate a summary. Which feature of Gemini 1.5 Pro is most suitable for this task?

A.Long context window (up to 1 million tokens)
B.Multimodal generation from text and images
C.Code generation and debugging
D.Function calling to retrieve external data
AnswerA

The long context allows ingesting entire video content for summarization.

Why this answer

Gemini 1.5 Pro's long context window of up to 1 million tokens allows it to process and analyze hour-long video content in a single pass, including both audio and visual frames. This capability is essential for generating a coherent summary of long-form video, as it can retain and reason over the entire video's context without needing to chunk or downsample the content.

Exam trap

The trap here is that candidates may confuse 'multimodal generation' (Option B) with the ability to process video, but Gemini 1.5 Pro's long context window is the specific feature designed for handling hour-long video content, not just the ability to handle multiple data types.

How to eliminate wrong answers

Option B is wrong because multimodal generation from text and images, while a feature of Gemini 1.5 Pro, does not specifically address the need to process hour-long video; it focuses on generating content from static images and text, not on handling long-duration video streams. Option C is wrong because code generation and debugging are unrelated to video analysis and summarization; they target software development tasks. Option D is wrong because function calling to retrieve external data is used to integrate with external APIs or databases, not for processing or summarizing video content directly.

Full explanation →
262
MCQmedium

During a RAG pipeline implementation, the retrieval system frequently returns irrelevant documents, causing the generator to produce incorrect answers. Which change is most likely to improve the relevance of retrieved documents?

A.Add a re-ranking step using a cross-encoder model to refine the top results.
B.Increase the number of documents retrieved from the vector store.
C.Use a different embedding model with higher vector dimension.
D.Decrease the chunk size of documents to reduce noise.
AnswerA

Re-ranking directly improves document relevance by deep semantic matching.

Why this answer

Re-ranking with a cross-encoder model evaluates query-document pairs more deeply, improving precision at the cost of latency. Increasing the number of documents may introduce more noise. Changing embedding dimension or chunk size may help but are less targeted.

Full explanation →
263
MCQhard

You are a machine learning engineer at a healthcare startup. Your team has developed a generative AI model that summarizes patient medical records. The model is deployed on Vertex AI Endpoints using a custom container. You have configured the endpoint with a single n1-standard-4 machine (4 vCPUs, 15 GB memory) without accelerators. The model uses a small transformer architecture. During load testing with 50 concurrent requests, you observe that the average latency is 8 seconds, which exceeds the requirement of 2 seconds. Additionally, some requests time out after 10 seconds. You suspect the CPU is the bottleneck. You also notice that the model inference code uses TensorFlow but is not optimized for inference. Which action should you take to reduce latency?

A.Reduce the model size by pruning and quantization, then redeploy.
B.Increase the request timeout to 30 seconds to accommodate the latency.
C.Enable autoscaling to handle the load with multiple replicas.
D.Deploy the model on a machine with a GPU and use TensorRT for inference optimization.
AnswerD

GPU acceleration and model optimization can drastically reduce latency.

Why this answer

The CPU is identified as the bottleneck, and deploying on a GPU with TensorRT optimization directly addresses this by accelerating the TensorFlow inference. TensorRT optimizes the model graph and fuses layers, significantly reducing latency for transformer-based models, which is essential to meet the 2-second requirement.

Exam trap

The trap here is that candidates may choose autoscaling (Option C) thinking it handles high concurrency, but they overlook that the per-request latency remains unchanged on CPU, failing to meet the 2-second requirement.

How to eliminate wrong answers

Option A is wrong because pruning and quantization reduce model size and can improve latency, but they may degrade model accuracy and do not address the fundamental CPU bottleneck as effectively as GPU acceleration with TensorRT. Option B is wrong because increasing the timeout to 30 seconds does not reduce latency; it only masks the problem, and requests still exceed the 2-second requirement, leading to poor user experience. Option C is wrong because autoscaling adds more replicas to handle concurrent requests, but each request still runs on a CPU-bound n1-standard-4 machine, so the per-request latency remains high and does not solve the CPU bottleneck.

Full explanation →
264
MCQmedium

A machine learning engineer is building a text-to-image model using Vertex AI. They want to reduce inference latency. Which strategy is most effective?

A.Use a larger image resolution
B.Use a smaller model variant
C.Enable batch processing
D.Increase the number of inference steps
AnswerB

Smaller models are faster.

Why this answer

Using a smaller model variant directly reduces the number of parameters and computational operations required per inference pass, which lowers latency. In text-to-image models like Imagen or Stable Diffusion, the model size is the primary driver of forward-pass time, so a smaller variant (e.g., fewer layers or reduced latent dimensions) yields faster generation.

Exam trap

The trap here is that candidates confuse throughput optimization (batch processing) with latency reduction, or assume that more steps or higher resolution improve quality without considering the latency trade-off.

How to eliminate wrong answers

Option A is wrong because larger image resolution increases the pixel space the model must process, which increases computational load and latency, not reduces it. Option C is wrong because batch processing improves throughput (images per second) but does not reduce per-request latency; it may even increase the time to first token for an individual request. Option D is wrong because increasing the number of inference steps (e.g., diffusion denoising steps) directly increases the sequential computation time, making latency worse.

Full explanation →
265
MCQeasy

A startup wants to quickly integrate a generative AI chatbot into their customer support platform. They need a solution that can answer questions based on their internal knowledge base with minimal setup. Which Google Cloud service should they use?

A.Use Model Garden to deploy a pre-built Q&A model
B.Call the Gemini API directly and implement grounding logic manually
C.Use Cloud AI Notebooks to fine-tune a model on their knowledge base
D.Use Vertex AI Agent Builder to create a conversational agent grounded in their data
AnswerD

Agent Builder offers pre-built components for grounding and conversation flow, enabling rapid deployment.

Why this answer

Vertex AI Agent Builder (now part of Vertex AI Agent Platform) provides a low-code, out-of-the-box solution for building conversational agents that are grounded in enterprise data. It automatically handles retrieval-augmented generation (RAG) by indexing the startup's internal knowledge base into a vector store and orchestrating the LLM to answer questions using only that data, requiring minimal setup compared to manual integration or fine-tuning.

Exam trap

The trap here is that candidates often confuse Model Garden (a model deployment hub) with a full conversational AI platform, overlooking that Vertex AI Agent Builder provides the essential grounding and orchestration layer that Model Garden lacks.

How to eliminate wrong answers

Option A is wrong because Model Garden is a model hub for discovering and deploying pre-built models, but it does not include built-in grounding logic or a conversational agent framework; deploying a Q&A model from Model Garden would still require custom code to connect it to the knowledge base. Option B is wrong because calling the Gemini API directly and implementing grounding logic manually requires significant custom development for document retrieval, chunking, embedding, and orchestration, which contradicts the 'minimal setup' requirement. Option C is wrong because Cloud AI Notebooks are designed for custom model training and fine-tuning, which is overkill and time-consuming for simply answering questions from an existing knowledge base; fine-tuning also risks catastrophic forgetting and does not natively support grounding in dynamic data.

Full explanation →
266
MCQhard

A company is deploying a chatbot that uses a foundation model. They want to minimize latency for user queries. Which action is most effective?

A.Use a larger model with more parameters
B.Disable safety filters
C.Increase the number of tokens
D.Use a smaller distilled model
AnswerD

Distilled models are optimized for speed.

Why this answer

Distilled models are smaller, faster versions of larger foundation models, trained to mimic their behavior while requiring fewer computational resources. This directly reduces inference latency because fewer parameters mean faster forward passes through the network, which is critical for real-time chatbot responses.

Exam trap

Google Cloud often tests the misconception that 'bigger is better' for performance, but in latency-constrained scenarios, model size is inversely related to speed, and candidates may overlook distillation as a standard optimization technique.

How to eliminate wrong answers

Option A is wrong because larger models with more parameters increase computational complexity and memory bandwidth requirements, which actually increases latency rather than reducing it. Option B is wrong because disabling safety filters does not affect model inference speed; safety filters are post-processing steps that add negligible latency compared to the model itself. Option C is wrong because increasing the number of tokens (the output length) forces the model to perform more autoregressive generation steps, which linearly increases latency per additional token.

Full explanation →
267
MCQeasy

A marketing team wants to generate social media posts in a consistent brand voice. They have a few examples of high-performing posts. Which prompt engineering technique should they use?

A.Few-shot prompting with examples of past successful posts
B.Fine-tuning the model on all past social media posts
C.Zero-shot prompting with a detailed description of the brand voice
D.Chain-of-thought prompting to explain reasoning
AnswerA

Few-shot examples guide the model to match the desired style and tone.

Why this answer

Few-shot prompting provides the model with examples of the desired output style and tone, enabling consistent brand voice without fine-tuning.

Full explanation →
268
MCQeasy

A data scientist wants to fine-tune a foundation model from Vertex AI Model Garden on their custom dataset. They want to choose a cost-effective method that updates only a small subset of parameters. Which fine-tuning approach should they use?

A.Full fine-tuning
B.Prompt tuning
C.Parameter-Efficient Fine-Tuning (PEFT) like LoRA
D.Reinforcement Learning from Human Feedback (RLHF)
AnswerC

PEFT methods update only a small subset of parameters.

Why this answer

Parameter-Efficient Fine-Tuning (PEFT) methods like LoRA (Low-Rank Adaptation) are specifically designed to update only a small subset of parameters (e.g., low-rank matrices injected into transformer layers) while keeping the majority of the foundation model frozen. This drastically reduces memory and compute costs compared to full fine-tuning, making it the most cost-effective choice for customizing a model from Vertex AI Model Garden on a custom dataset.

Exam trap

The trap here is that candidates often confuse prompt tuning (which does not update model parameters) with parameter-efficient fine-tuning (which updates a small subset of parameters), leading them to incorrectly select Option B as a cost-effective method for updating parameters.

How to eliminate wrong answers

Option A is wrong because full fine-tuning updates all model parameters, which is computationally expensive and memory-intensive, contradicting the requirement for a cost-effective method that updates only a small subset of parameters. Option B is wrong because prompt tuning is a soft-prompt technique that does not update any model parameters; instead, it learns a small set of virtual tokens prepended to the input, which is not a parameter-efficient fine-tuning method (it is a prompt-based approach). Option D is wrong because Reinforcement Learning from Human Feedback (RLHF) is a training paradigm that uses human preferences to align model behavior, typically requiring multiple models (reward model, policy model) and full or PEFT fine-tuning, and it is not primarily a cost-effective method for updating a small subset of parameters on a custom dataset.

Full explanation →
269
MCQmedium

A company is deploying a chatbot that must ensure customer data remains within the European Union. Which approach should they take?

A.Use Vertex AI Agent Builder with global endpoint
B.Use the Gemini API with a regional endpoint in europe-west4
C.Use Vertex AI with a multi-region endpoint
D.Deploy a custom model on GKE in a specific region
AnswerB

Regional endpoints ensure data remains in the specified region.

Why this answer

Using the Gemini API with a regional endpoint in europe-west4 ensures that all data processing and storage occurs within the European Union. Regional endpoints restrict data residency to the specified region, complying with GDPR and other EU data sovereignty requirements.

Exam trap

The trap here is that candidates often confuse multi-region endpoints (e.g., 'eu') with regional endpoints, assuming that any EU-based endpoint guarantees data stays within a specific country, when in fact multi-region endpoints can span multiple EU countries and may not meet strict single-country data residency requirements.

How to eliminate wrong answers

Option A is wrong because Vertex AI Agent Builder with a global endpoint does not guarantee data residency within the EU; global endpoints may route data to any Google Cloud region worldwide. Option C is wrong because Vertex AI with a multi-region endpoint (e.g., 'eu') can still process data across multiple EU regions, but it does not provide the strict single-region data residency guarantee that a regional endpoint offers. Option D is wrong because deploying a custom model on GKE in a specific region only controls compute location, not the underlying AI service endpoints or data processing paths, which may still involve global services or cross-region data transfer.

Full explanation →
270
MCQmedium

A data scientist wants to share a Jupyter notebook with interactive visualizations for a Kaggle competition. Which tool should they use to easily publish and collaborate?

A.Google Colab
B.Kaggle Notebooks
C.Vertex AI Workbench
D.Cloud Shell Editor
AnswerB

Kaggle Notebooks are integrated with Kaggle datasets and competitions.

Why this answer

B is correct because Kaggle Notebooks are natively integrated with the Kaggle competition platform, allowing data scientists to publish interactive visualizations directly within the competition environment and collaborate with other participants. Unlike generic tools, Kaggle Notebooks provide seamless access to competition datasets, GPU/TPU resources, and version control tailored for Kaggle workflows.

Exam trap

The Generative AI Leader exam often tests the distinction between general-purpose tools (like Colab) and platform-specific tools (like Kaggle Notebooks), trapping candidates who overlook the tight integration with competition datasets and submission APIs.

How to eliminate wrong answers

Option A is wrong because Google Colab is a general-purpose cloud-based Jupyter notebook environment that lacks native integration with Kaggle competition datasets and submission systems, requiring manual uploads and API workarounds. Option C is wrong because Vertex AI Workbench is an enterprise MLOps tool designed for production model development and deployment, not for lightweight sharing and collaboration on Kaggle competitions. Option D is wrong because Cloud Shell Editor is a lightweight code editor in the Google Cloud console intended for quick edits and terminal access, not for publishing interactive Jupyter notebooks with visualizations.

Full explanation →
271
MCQmedium

A global e-commerce company wants to translate product descriptions into 50 languages with high accuracy. They need to handle domain-specific terms (e.g., 'size chart', 'return policy'). Which approach should they use?

A.Use the Gemini API with a prompt like 'Translate to French'
B.Build a custom agent with Vertex AI Agent Builder
C.Use Vertex AI Translation with custom glossaries
D.Use Imagen to generate translated images
AnswerC

Custom glossaries ensure domain-specific terms are translated correctly.

Why this answer

Vertex AI Translation with custom glossaries is specifically designed for high-accuracy, domain-specific translations. Custom glossaries allow you to define precise translations for terms like 'size chart' and 'return policy', ensuring consistency across 50 languages. This approach leverages Google's neural machine translation models while overriding generic translations with your business-specific terminology.

Exam trap

The trap here is that candidates may confuse general-purpose generative AI APIs (like Gemini) with specialized translation services, or assume that any AI model can handle domain-specific translation without customization, when in fact glossaries are required for consistent, accurate terminology.

How to eliminate wrong answers

Option A is wrong because the Gemini API is a general-purpose generative AI model, not a specialized translation service; it lacks built-in support for custom glossaries and may produce inconsistent or hallucinated translations for domain-specific terms. Option B is wrong because Vertex AI Agent Builder is designed for building conversational agents and workflows, not for bulk, high-accuracy translation tasks; it would require significant custom development to replicate glossary-based translation. Option D is wrong because Imagen is a text-to-image generation model, not a translation tool; it cannot translate text and would be irrelevant for translating product descriptions.

Full explanation →
272
MCQmedium

A developer is using Vertex AI Studio to design a prompt for a content moderation system. They need the model to return a structured JSON with fields 'category' and 'confidence_score'. Which prompt engineering technique should they use?

A.Set a role prompt instructing the model to act as a content moderator
B.Use a response schema (structured output) in Vertex AI Studio
C.Use zero-shot prompting and rely on the model's ability to infer JSON
D.Include a few-shot example of the desired JSON output in the prompt
AnswerB

Response schemas enforce the JSON format and fields, ensuring consistent structured output from the model.

Why this answer

Response schemas in Vertex AI Studio allow you to define the exact structure of the output, ensuring consistent JSON. Few-shot provides examples but not a strict schema. Zero-shot is too variable.

Role-setting is for tone, not structure.

Full explanation →
273
MCQmedium

A company is using a large language model for a customer-facing chat application. They notice that the model sometimes generates plausible-sounding but incorrect information. Which strategy is most effective to reduce this issue?

A.Fine-tune the model on a dataset of corrected conversations.
B.Reduce the context window to limit the information the model considers.
C.Implement Retrieval-Augmented Generation (RAG) to retrieve relevant documents before generating a response.
D.Increase the temperature parameter to make the model more deterministic.
AnswerC

RAG grounds the model's output in retrieved evidence, directly reducing hallucinations.

Why this answer

RAG grounds responses by retrieving relevant, verified information from a trusted source, reducing hallucinations. Fine-tuning on incorrect data would not help; increasing temperature increases randomness; reducing context window may lose relevant info.

Full explanation →
274
Multi-Selectmedium

A company is choosing between Gemini Pro and Gemini Ultra for a document summarization task. Which THREE factors should they consider when deciding between the two model variants? (Choose THREE)

Select 3 answers
A.Model capability and accuracy
B.Availability in Google AI Studio
C.Latency requirements
D.Cost per token
E.Multimodal support
AnswersA, C, D

Ultra is more capable and accurate for complex tasks, but Pro may suffice for simpler summarization.

Why this answer

Model capability and accuracy (A) is correct because Gemini Pro is optimized for high-throughput, cost-efficient tasks with lower accuracy demands, while Gemini Ultra is designed for complex, high-accuracy reasoning. The choice directly impacts the quality of the summarization output, as Ultra uses a larger parameter count and more advanced attention mechanisms to handle nuanced context, whereas Pro may struggle with ambiguous or lengthy documents.

Exam trap

The Generative AI Leader exam often tests the misconception that model availability or multimodal support are primary selection criteria, when in fact the core trade-offs are capability, latency, and cost per token for the specific task.

Full explanation →
275
MCQhard

A research team is using Imagen to generate images for a marketing campaign. They notice that the generated images sometimes contain distorted faces or unnatural object placements. They want to improve consistency without sacrificing image diversity. Which approach should they try first?

A.Reduce the top-p value (e.g., from 1.0 to 0.9).
B.Increase the temperature to 1.5
C.Switch from Imagen to Gemini Pro for image generation
D.Set top-k to 1
AnswerA

Lowering top-p cuts off the tail of unlikely tokens, reducing artifacts while still allowing a range of plausible tokens for diversity.

Why this answer

Adjusting top-p (nucleus sampling) reduces the set of possible tokens to the most probable ones, which often removes low‑probability artifacts while preserving diversity. Prompt engineering and fine‑tuning can also help but are more involved. Temperature reduction may lower diversity too much.

Full explanation →
276
MCQhard

A large enterprise runs a production application that uses the Gemini API on Vertex AI for real-time content moderation. They are experiencing occasional 429 (Too Many Requests) errors during peak hours. Their current quota is 1000 requests per minute (RPM) and they are hitting around 950 RPM on average, with spikes up to 1050. They have already implemented exponential backoff and retry logic. They need to reduce the error rate without reducing the quality of moderation. Which additional measure should they take?

A.Deploy the model on a dedicated Vertex AI endpoint with autoscaling.
B.Switch to a lower-tier model like Gemini 1.0 Pro to reduce quota consumption.
C.Implement a local caching layer for common moderation queries.
D.Request a quota increase from Google Cloud support.
AnswerC

Caching eliminates duplicate requests, reducing the request rate and errors.

Why this answer

Implementing a local caching layer for common moderation queries reduces the number of identical requests sent to the Gemini API, directly lowering the effective RPM without compromising moderation quality. Since the enterprise is already using exponential backoff and retry logic, caching addresses the root cause of hitting quota limits by eliminating redundant API calls, which is a standard pattern for rate-limit mitigation in production AI workloads.

Exam trap

Google Cloud often tests the misconception that scaling infrastructure (Option A) or switching models (Option B) solves API quota issues, when the real constraint is the API rate limit itself, which requires reducing the number of calls through caching or other client-side optimizations.

How to eliminate wrong answers

Option A is wrong because deploying a dedicated endpoint with autoscaling does not increase the quota limit; it only scales compute resources, and the 429 errors are due to API quota exhaustion, not endpoint capacity. Option B is wrong because switching to a lower-tier model like Gemini 1.0 Pro would reduce quality of moderation, which violates the requirement to not reduce quality, and it does not address the fundamental issue of hitting the RPM quota. Option D is wrong because requesting a quota increase is a valid long-term solution but does not address the immediate need to reduce error rate without reducing quality; it also assumes Google Cloud will approve the increase, which is not guaranteed, and it does not optimize existing usage.

Full explanation →
277
MCQeasy

What is the primary purpose of SynthID, developed by Google DeepMind?

A.To improve the factual accuracy of AI-generated text
B.To automatically filter harmful content in AI outputs
C.To embed an invisible watermark in AI-generated content for identification
D.To reduce bias in training datasets
AnswerC

SynthID watermarks help trace the origin of AI-generated images, audio, or text.

Why this answer

SynthID is designed to embed invisible watermarks into AI-generated content to help identify its origin.

Full explanation →
278
MCQmedium

A manufacturing company wants to use generative AI to create maintenance manuals from sensor data. The manuals must be accurate and reflect the latest equipment configurations. Which approach best ensures data freshness and consistency?

A.Train the model in real-time as sensor data streams in.
B.Periodically retrain the model with the latest sensor data.
C.Have human technicians review and update the manuals manually.
D.Use a retrieval-augmented generation (RAG) system that queries a live database of sensor configurations.
AnswerD

RAG ensures responses are based on the most current data.

Why this answer

A retrieval-augmented generation (RAG) system retrieves the most current equipment configurations directly from a live database at inference time, ensuring the generated manual reflects real-time sensor data without requiring model retraining. This approach decouples the static knowledge in the LLM from the dynamic data source, guaranteeing both accuracy and freshness while avoiding the latency and cost of continuous retraining.

Exam trap

Google Cloud often tests the misconception that retraining (Option B) is the only way to keep an LLM current, when in fact RAG provides a more efficient and accurate mechanism for incorporating live data without modifying the model itself.

How to eliminate wrong answers

Option A is wrong because training a model in real-time as sensor data streams in is impractical due to catastrophic forgetting, high computational overhead, and the inability of online learning to guarantee that the model's weights stabilize to reflect the latest configurations without extensive validation. Option B is wrong because periodic retraining introduces a window of staleness between retraining cycles, during which sensor data may change, leading to manuals that are not current; it also requires significant infrastructure for data collection, preprocessing, and model deployment. Option C is wrong because manual review and update by human technicians is slow, error-prone, and cannot scale to the volume and velocity of sensor data, defeating the purpose of using generative AI for automation.

Full explanation →
279
MCQmedium

A company wants to use Generative AI for customer support chatbots. They are concerned about cost and latency. Which deployment option best balances these concerns?

A.Deploy an open-source model on-premise to avoid cloud costs
B.Rely on a third-party chatbot API that abstracts the model
C.Use the largest available foundation model via API for highest accuracy
D.Use a fine-tuned version of a smaller model on Vertex AI with response caching
AnswerD

A tuned smaller model reduces compute cost and caching minimizes repeated inference, lowering latency. Vertex AI provides scalable infrastructure.

Why this answer

Using a fine-tuned smaller model on Vertex AI with response caching reduces both cost and latency. Smaller models require fewer computational resources, and caching avoids redundant inference calls, directly addressing the company's concerns without sacrificing accuracy for the specific task.

Exam trap

Google Cloud often tests the misconception that 'larger model = better accuracy always' or that 'on-premise is always cheaper,' ignoring the total cost of ownership, scaling overhead, and the efficiency gains from fine-tuning and caching for specific use cases.

How to eliminate wrong answers

Option A is wrong because deploying on-premise incurs high upfront hardware and maintenance costs, and may not scale efficiently for variable customer support loads, often increasing total cost of ownership (TCO) despite avoiding cloud fees. Option B is wrong because relying on a third-party chatbot API abstracts the model but does not inherently optimize cost or latency; it may introduce per-call pricing and network overhead, and the provider controls model size and caching. Option C is wrong because using the largest available foundation model via API maximizes accuracy but also maximizes inference cost and latency due to higher parameter count and compute requirements, which is the opposite of balancing cost and latency.

Full explanation →
280
MCQmedium

A company is evaluating the ROI of deploying a GenAI-powered code review assistant. Which metric would BEST capture the quality improvement from using the assistant?

A.Number of code reviews completed per week
B.Lines of code written per developer per day
C.Developer satisfaction survey score
D.Defect escape rate (bugs found in production)
AnswerD

Defect escape rate directly reflects the quality of code reviews; fewer production bugs indicate higher quality.

Why this answer

Defect escape rate measures bugs that reach production; a reduction indicates higher code quality. Productivity metrics like lines of code per hour measure speed, not quality. User acceptance is a satisfaction measure, not direct quality.

Full explanation →
281
Multi-Selecthard

A company is developing a generative AI application that will be used by customers in the EU. To comply with the GDPR and the upcoming EU AI Act, which THREE measures should they implement? (Select 3)

Select 3 answers
A.Store and process all user data within the EU or in regions with adequacy decisions
B.Use only open-source models to ensure transparency
C.Automatically delete all user logs after 30 days
D.Provide users with the ability to obtain an explanation of the AI's decisions
E.Clearly inform users that they are interacting with an AI system
AnswersA, D, E

Data residency is a GDPR requirement for EU user data.

Why this answer

GDPR requires data residency/local processing (measures 1), explanation of AI decisions (right to explanation) (measure 3), and transparency about AI interaction (measure 5). The other options are not mandated by these regulations.

Full explanation →
282
MCQhard

A company is evaluating the ROI of a generative AI project. Which metric is most appropriate?

A.Reduction in time to complete tasks using the generative AI tool
B.Reduction in model error rate on a test set
C.Increase in user satisfaction scores
D.Cost per inference compared to historical average
AnswerA

Time savings directly translate to labor cost reduction or increased throughput, providing a clear ROI.

Why this answer

The primary business justification for a generative AI project is operational efficiency, measured directly by the reduction in time to complete tasks. Unlike technical metrics such as model error rate, this metric ties the AI's output to tangible productivity gains, which is the core of ROI analysis in a business context. Generative AI tools are designed to augment human workflows, so time savings translate into cost savings and increased throughput, making it the most appropriate metric for evaluating return on investment.

Exam trap

Google Cloud often tests the distinction between technical performance metrics (like model error rate) and business outcome metrics, trapping candidates who default to evaluating AI models as they would in a data science context rather than from a business leadership perspective.

How to eliminate wrong answers

Option B is wrong because reduction in model error rate on a test set is a technical performance metric, not a business ROI metric; it measures model accuracy but does not account for the cost of deployment, user adoption, or actual business value generated. Option C is wrong because increase in user satisfaction scores, while valuable, is a lagging indicator that can be influenced by factors unrelated to the AI's direct impact on productivity or cost, and it does not quantify financial return. Option D is wrong because cost per inference compared to historical average focuses solely on operational cost efficiency, ignoring the revenue or time-saving benefits that the generative AI tool provides, thus failing to capture the full ROI picture.

Full explanation →
283
MCQhard

An enterprise uses a fine-tuned PaLM 2 model for code generation. They want to ensure the generated code passes security audits. Which combination of techniques would be most effective?

A.Integrate a static analysis tool in the pipeline and add a safety filter to reject code containing dangerous functions.
B.Use a few-shot prompt with examples of secure code and set temperature to 1.0.
C.Fine-tune the model on a dataset of insecure code and use top-p=0.9.
D.Increase the model's context window and use a system instruction to 'be secure'.
AnswerA

Static analysis and safety filters directly block insecure code patterns.

Why this answer

Integrating a static analysis tool (e.g., SonarQube, Checkmarx) into the pipeline provides automated, rule-based scanning for security vulnerabilities like SQL injection or buffer overflows, while a safety filter explicitly blocks generated code containing dangerous functions (e.g., eval(), exec()). This combination creates a defense-in-depth approach that catches both known vulnerability patterns and explicitly prohibited operations, which is essential for passing security audits.

Exam trap

The Generative AI Leader exam often tests the misconception that prompt engineering alone (e.g., system instructions or few-shot examples) is sufficient for security, when in fact deterministic validation and filtering techniques are required to enforce constraints reliably.

How to eliminate wrong answers

Option B is wrong because using a few-shot prompt with secure code examples does not guarantee the model will consistently avoid generating insecure code—temperature=1.0 increases randomness, making the output less deterministic and more likely to deviate from the secure examples. Option C is wrong because fine-tuning on a dataset of insecure code would teach the model to generate vulnerable patterns, which is counterproductive for security; top-p=0.9 does not prevent the model from outputting those learned insecure constructs. Option D is wrong because increasing the context window and using a system instruction to 'be secure' provides no enforcement mechanism—the model can still generate insecure code if the instruction is not followed, and there is no validation step to catch violations.

Full explanation →
284
Multi-Selecthard

Which THREE factors should be considered when choosing between Gemini 1.5 Pro and Gemini 1.5 Flash for a customer-facing chatbot? (Choose three.)

Select 3 answers
A.Cost constraints: Flash is more cost-effective per token
B.Task complexity: Pro is better for complex reasoning
C.Safety filters: Pro has stricter safety defaults
D.Latency requirements: Flash provides faster responses
E.Multimodal capability: Flash does not support image input
AnswersA, B, D

Flash is cheaper.

Why this answer

Gemini 1.5 Flash is designed as a cost-optimized model, offering significantly lower per-token pricing compared to Gemini 1.5 Pro. For a customer-facing chatbot with high query volumes, cost efficiency is a primary consideration, making Flash the more economical choice for routine interactions.

Exam trap

The trap here is that candidates often assume Flash lacks multimodal capabilities or that Pro has stricter safety defaults, when in fact both models share the same safety configuration and both support multimodal inputs, with the key differentiators being cost, latency, and task complexity.

Full explanation →
285
MCQmedium

A company wants to use a generative AI model to create marketing content that will be used commercially. They are concerned about copyright infringement from the model's training data. Which step should they take to mitigate this risk?

A.Use SynthID to watermark the content
B.Add a disclaimer that content is AI-generated
C.Review the model's training data provenance and ensure the model is trained on properly licensed or public domain data
D.Use only open-source models
AnswerC

Understanding data provenance helps assess copyright risk.

Why this answer

Reviewing the training data provenance and using models with appropriate licenses is critical. The other options either do not address the root cause or are insufficient.

Full explanation →
286
Multi-Selectmedium

A company wants to implement a change management strategy for GenAI adoption. Which TWO actions are MOST effective in driving adoption among employees?

Select 2 answers
A.Remove existing tools to force adoption
B.Appoint AI champions in each department to advocate and assist
C.Limit access to GenAI tools to senior leadership only
D.Make the use of GenAI tools mandatory for all tasks
E.Provide comprehensive training programs for employees
AnswersB, E

Champions provide local support and motivation.

Why this answer

Training programs build competence; AI champions provide peer support. Mandatory adoption and tool removal are counterproductive.

Full explanation →
287
Multi-Selecteasy

A company is considering using gen AI for customer support. Which two business strategies are most important for success?

Select 2 answers
A.Measure customer satisfaction metrics
B.Ignore data privacy
C.Deploy without testing
D.Ensure human-in-the-loop for critical interactions
E.Use the cheapest model
AnswersA, D

Metrics help evaluate success and guide improvements.

Why this answer

Measuring customer satisfaction metrics (A) is critical because it provides quantitative feedback on the generative AI system's performance, enabling iterative improvements to the model's responses and alignment with business goals. Without metrics like CSAT or NPS, the company cannot validate whether the AI is reducing resolution time or improving user experience, which are key ROI indicators for gen AI deployments.

Exam trap

Google Cloud often tests the misconception that cost optimization (cheapest model) or speed-to-market (deploy without testing) are primary success factors, when in reality governance, safety, and continuous measurement are the foundational strategies for sustainable gen AI adoption.

Full explanation →
288
MCQmedium

A financial analyst wants to run a regression model inside BigQuery using SQL, without moving data to a separate ML environment. Which Google Cloud service allows this directly?

A.BigQuery ML
B.AI Platform
C.AutoML Tables
D.Vertex AI
AnswerA

BigQuery ML allows creating and running ML models using SQL directly in BigQuery.

Why this answer

BigQuery ML enables creating and running ML models using standard SQL queries directly in BigQuery. Vertex AI and AI Platform require data to be moved, and AutoML is a higher-level service but still not SQL-based.

Full explanation →
289
Multi-Selecteasy

A developer wants to use Google Cloud generative AI to build a multimodal application that can answer questions about images and text. Which TWO services are most appropriate?

Select 2 answers
A.Chirp
B.Vertex AI
C.Imagen
D.Codey
E.Gemini Pro Vision
AnswersB, E

Vertex AI provides a unified platform to deploy and serve models like Gemini for multimodal applications.

Why this answer

Vertex AI is correct because it provides a unified platform for building, deploying, and managing multimodal generative AI models, including support for both image and text inputs. It offers access to foundational models like Gemini Pro Vision, enabling developers to create applications that answer questions about images and text through a single API endpoint.

Exam trap

The trap here is that candidates often confuse specialized single-modal services (like Imagen for image generation or Codey for code) with the multimodal platform (Vertex AI) that integrates multiple capabilities, leading them to select a single-purpose service instead of the correct platform and its multimodal model.

Full explanation →
290
Multi-Selectmedium

An enterprise wants to use generative AI to help employees search through internal documents, including text, scanned PDFs, and images. They need to index the content and enable semantic search. Which TWO Google Cloud services should they use? (Choose two.)

Select 2 answers
A.Document AI
B.Vertex AI Vector Search
C.Vertex AI Embeddings API
D.Cloud Storage with Object Versioning
E.Gemini API for content generation
AnswersB, C

Performs similarity search over embeddings for semantic retrieval.

Why this answer

Vertex AI Embeddings API generates vector embeddings from text and images. Vertex AI Vector Search allows semantic search over those embeddings. Document AI is for document processing but not for embeddings or search.

Gemini API is for generation, not indexing. Cloud Storage is for storage, not search.

Full explanation →
291
MCQhard

A financial services firm wants to deploy generative AI for automated investment advice. They are subject to strict regulatory oversight requiring explainability and audit trails. Which strategy best meets these requirements?

A.Fine-tune a model on historical trading data without human review.
B.Use a black-box large language model with monitoring.
C.Deploy a rule-based system augmented with generative AI for content generation.
D.Implement human-in-the-loop with full logging of model inputs, outputs, and human decisions.
AnswerD

This provides a transparent audit trail and human accountability, satisfying regulatory demands.

Why this answer

It directly addresses the regulatory requirements for explainability and audit trails by incorporating human oversight and comprehensive logging. The human-in-the-loop (HITL) mechanism ensures that critical investment decisions are reviewed by qualified professionals, while full logging of model inputs, outputs, and human decisions creates a transparent, auditable record. This approach satisfies financial regulations like MiFID II or SEC rules that mandate explainability and accountability in automated advice systems.

Exam trap

Google Cloud often tests the misconception that monitoring or rule-based augmentation alone is sufficient for regulatory compliance, when in fact strict oversight and complete audit trails are mandatory for explainability in high-stakes domains like finance.

How to eliminate wrong answers

Option A is wrong because fine-tuning a model on historical trading data without human review introduces risks of overfitting to past market conditions and lacks the necessary audit trail and explainability for regulatory compliance. Option B is wrong because using a black-box large language model with monitoring still fails to provide the required explainability, as the internal decision-making process remains opaque and cannot be audited or justified to regulators. Option C is wrong because a rule-based system augmented with generative AI for content generation, while more transparent, still lacks the structured human oversight and full logging of decisions needed to meet strict audit trail requirements, and the generative AI component can introduce unpredictable outputs that undermine explainability.

Full explanation →
292
MCQmedium

An enterprise deploys a large language model (LLM) for internal document summarization. Users complain that summaries sometimes include statements not present in the original document. Which mitigation strategy should the team prioritize to address this hallucination issue?

A.Train a discriminator model to detect hallucinations and perform adversarial training.
B.Implement retrieval-augmented generation (RAG) to ground the model in the original documents and require citations.
C.Apply reinforcement learning from human feedback (RLHF) using a reward model that penalizes hallucinations.
D.Reduce the model's temperature parameter to 0 to make outputs deterministic.
AnswerB

RAG ties outputs to source documents, reducing hallucinations.

Why this answer

Retrieval-Augmented Generation (RAG) is the most direct and effective mitigation for hallucination in document summarization because it forces the LLM to base its output on retrieved chunks of the original document. By requiring citations, the model must reference specific passages, making it verifiable and reducing the likelihood of fabricating content. This grounds the generation in the source material, addressing the root cause of hallucination—lack of factual grounding—rather than relying on post-hoc correction or output tuning.

Exam trap

Google Cloud often tests the misconception that reducing temperature or applying RLHF alone can solve hallucination, when in fact these methods do not provide the explicit grounding that RAG offers for document-specific tasks.

How to eliminate wrong answers

Option A is wrong because training a discriminator model and performing adversarial training is a complex, resource-intensive approach that does not directly prevent hallucinations at inference time; it only improves robustness against adversarial inputs, not factual grounding. Option C is wrong because RLHF with a reward model that penalizes hallucinations can reduce their frequency over time, but it requires extensive human feedback and fine-tuning, and does not guarantee grounding in specific source documents for each summary. Option D is wrong because reducing the temperature parameter to 0 makes outputs deterministic but does not eliminate hallucinations—it only reduces randomness; the model can still confidently generate false statements that were never in the source document.

Full explanation →
293
Multi-Selecteasy

A company wants to start using Gemini for Google Workspace to improve productivity. Which TWO applications can provide meeting summarization features? (Choose 2)

Select 2 answers
A.Google Docs
B.Google Chat
C.Google Sheets
D.Gmail
E.Google Meet
AnswersB, E

Chat can display meeting summaries if integrated with Meet.

Why this answer

Google Meet and Google Chat both offer meeting summarization features powered by Gemini. In Google Meet, Gemini can automatically generate meeting notes and summaries after a recorded meeting. In Google Chat, Gemini can summarize conversations and meetings that occurred within a space, providing a concise recap of key points and action items.

Exam trap

The trap here is that candidates may confuse general AI assistance features (like drafting emails or documents) with the specific meeting summarization capability, which is only available in Google Meet and Google Chat.

Full explanation →
294
Multi-Selectmedium

A data scientist is using Vertex AI to build a question-answering system. They want to use the Responsible AI toolkit to document and communicate the model's characteristics. Which TWO tools from the toolkit are MOST relevant? (Select two.)

Select 2 answers
A.Cloud DLP
B.TensorBoard
C.Model Cards
D.Datasheets for Datasets
E.Vertex AI Workbench
AnswersC, D

Model Cards provide standardized documentation of model behavior.

Why this answer

Model Cards document model performance, intended use, and limitations. Datasheets for Datasets document dataset creation and composition. Both are part of the Responsible AI toolkit for transparency.

Full explanation →
295
MCQmedium

A healthcare provider plans to implement gen AI for clinical note summarization. They have limited AI expertise. Which Google Cloud approach best aligns with their business strategy?

A.Hire a team of data scientists
B.Use Vertex AI Agent Builder with pre-built templates
C.Deploy an open-source model on Compute Engine
D.Build a custom model from scratch
AnswerB

Leverages managed services and reduces the need for in-house AI expertise.

Why this answer

Vertex AI Agent Builder provides pre-built templates and a low-code interface specifically designed for organizations with limited AI expertise. It enables rapid deployment of generative AI solutions like clinical note summarization without requiring deep data science skills, directly aligning with the healthcare provider's business strategy of minimizing technical overhead while leveraging AI.

Exam trap

Google Cloud often tests the misconception that 'more technical control' (e.g., custom models or open-source deployment) is always better, but the trap here is that the question explicitly prioritizes business strategy and limited expertise, making low-code/no-code solutions like Vertex AI Agent Builder the correct choice over technically complex alternatives.

How to eliminate wrong answers

Option A is wrong because hiring a team of data scientists contradicts the 'limited AI expertise' constraint and introduces significant cost and time overhead, which is not a strategic fit for rapid implementation. Option C is wrong because deploying an open-source model on Compute Engine requires substantial DevOps, model tuning, and infrastructure management expertise, which the provider lacks. Option D is wrong because building a custom model from scratch demands advanced machine learning skills, large labeled datasets, and extensive training resources, making it impractical for an organization with limited AI expertise.

Full explanation →
296
Multi-Selectmedium

A machine learning engineer wants to reduce the latency of a Gemini-based chatbot running in production. Which TWO strategies would be MOST effective?

Select 2 answers
A.Switch from Gemini Pro to Gemini Flash
B.Reduce the max_output_tokens parameter
C.Enable streaming mode
D.Fine-tune the model on the specific task
E.Increase the temperature to 1.0
AnswersA, B

Flash is optimized for lower latency and cost compared to Pro.

Why this answer

Gemini Flash is a lighter, more efficient model variant designed for lower latency and higher throughput compared to Gemini Pro. By switching to Flash, the engineer reduces the computational overhead per request, directly decreasing response time for the chatbot.

Exam trap

A common misconception in Google exams is that streaming reduces total latency, but it only improves perceived latency (time-to-first-token) while total processing time remains unchanged.

Full explanation →
297
Multi-Selectmedium

Which TWO techniques are most effective for improving factual accuracy in a generative AI model's responses? (Choose two.)

Select 2 answers
A.Retrieval-Augmented Generation (RAG) with curated datasets.
B.Increasing the model's temperature to 1.5.
C.Grounding with a trusted knowledge base.
D.Using longer system prompts with multiple instructions.
E.Fine-tuning on a large corpus of general text.
AnswersA, C

RAG retrieves relevant, up-to-date documents to inform responses.

Why this answer

Grounding and RAG both provide external authoritative sources to enhance factual accuracy. Fine-tuning on general data doesn't guarantee accuracy, and increasing temperature hurts accuracy. Prompt engineering is helpful but not as robust as retrieval-based methods.

Full explanation →
298
Multi-Selectmedium

A machine learning engineer is using a generative AI model that occasionally outputs biased content. Which TWO actions should they take to address this issue?

Select 2 answers
A.Use a single test set for simplicity
B.Remove all safety filters to allow more freedom in outputs
C.Adjust the training data to be more representative of the target population
D.Increase the model's size to improve generalization
E.Evaluate the model outputs on diverse test sets to detect bias
AnswersC, E

More representative data reduces bias.

Why this answer

Evaluating bias with diverse test sets helps identify issues, and adjusting training data representativeness helps mitigate bias at the source.

Full explanation →
299
MCQhard

A data scientist sees the above error when trying to deploy a model to an endpoint. What is the most likely cause?

A.The IAM permissions are insufficient
B.The model import into Vertex AI Model Registry is still in progress
C.The endpoint does not exist
D.The model is already deployed to another endpoint
AnswerB

Model is still DEPLOYING, not ready.

Why this answer

The error indicates that the model is not yet fully imported into the Vertex AI Model Registry. Deploying a model to an endpoint requires the model resource to be in an 'ACTIVE' state; if the import is still in progress, the deployment request will fail. This is a common timing issue when a model is uploaded but not yet registered.

Exam trap

Google Cloud often tests the misconception that any deployment failure is due to permissions or missing resources, when in fact the model's lifecycle state (e.g., still importing) is the root cause.

How to eliminate wrong answers

Option A is wrong because insufficient IAM permissions would typically result in a 403 Forbidden error, not a model-not-found or import-in-progress error. Option C is wrong because if the endpoint did not exist, the error would be a 404 Not Found for the endpoint resource, not a model import issue. Option D is wrong because a model can be deployed to multiple endpoints simultaneously; the error would instead mention a conflict or quota limit, not an import status.

Full explanation →
300
MCQeasy

A startup wants to quickly prototype a multimodal AI application that can process images and text using Gemini. They have minimal budget and need a free tier for initial development. Which access tier should they use?

A.Vertex AI
B.Google AI Studio
C.Cloud TPU
D.TensorFlow Hub
AnswerB

Google AI Studio offers a free tier with rate limits, perfect for experimentation and prototyping before moving to production.

Why this answer

Google AI Studio is the correct choice because it offers a free tier specifically designed for rapid prototyping with Gemini models, including multimodal capabilities for processing images and text. It provides a web-based interface and API access without requiring a billing account, making it ideal for startups with minimal budget. Vertex AI, while powerful, requires a paid Google Cloud project and is intended for production deployment, not free-tier prototyping.

Exam trap

The trap here is that candidates often confuse Vertex AI's free trial credits (which still require a billing account) with a true free tier, or they assume TensorFlow Hub provides API access to Gemini, when in fact it only hosts static model artifacts for download.

How to eliminate wrong answers

Option A is wrong because Vertex AI is a managed ML platform that requires a billing-enabled Google Cloud project and charges for usage, making it unsuitable for a minimal-budget free-tier prototype. Option C is wrong because Cloud TPU is a hardware accelerator for training large models, not an access tier for using pre-built multimodal APIs like Gemini, and it incurs significant costs. Option D is wrong because TensorFlow Hub is a repository for reusable model components, not a service for accessing Gemini's multimodal capabilities, and it does not provide a free tier for API-based prototyping.

Full explanation →
Page 3

Page 4 of 10

Page 5

All pages

12345678910

Practice Generative AI Leader by domain

Target a specific domain to shore up weak areas.

Fundamentals of Generative AIBusiness Strategies for Generative AI SolutionsGoogle Cloud's Generative AI OfferingsTechniques to Improve Generative AI Model Output
See all domains with question counts →