Generative AI Leader

Question 1

A retail company wants to customize a large foundation model to consistently follow their internal product-naming conventions and writing style. Their ML team has a modest compute budget and wants to avoid the cost and time of retraining all of the model's parameters. They also want to keep the original base model intact so it can still serve other general tasks. Which customization approach best fits these constraints?

Accepted Answer

Adapter-based (parameter-efficient) tuning that trains a small set of additional parameters while leaving the base model weights unchanged

Answer

Full fine-tuning that updates all of the model's parameters on the company's dataset

Answer

Pretraining a new foundation model from scratch on the company's product catalog

Answer

Increasing the model's temperature setting so it adopts the company's writing style

Question 2

A large insurance company has deployed a gen AI assistant to help claims adjusters draft case summaries. Six months later, leadership finds that only 15% of adjusters use the tool regularly. Surveys reveal that many employees fear the tool will replace their jobs and are unsure how it fits into their existing workflow. Which action should the leadership team prioritize to improve adoption?

Accepted Answer

Launch a structured change management program with role-specific training and clear communication about how the tool augments adjusters' work

Answer

Fine-tune the underlying model on more claims data to improve summary accuracy

Answer

Mandate tool usage by tying it to individual performance reviews and quotas

Answer

Switch to a cheaper, faster foundation model to reduce operating costs

Question 3

A retail company deploys a generative AI assistant trained largely on historical customer support transcripts from one region. After launch, leaders notice the assistant frequently uses idioms and cultural references unfamiliar to customers in other regions, causing confusion. Which responsible AI concern does this situation most directly illustrate?

Accepted Answer

Bias, because skewed training data caused outputs that do not serve the broader customer base fairly

Answer

Hallucination, because the model invented references that do not exist

Answer

Privacy, because customer transcripts were used in training

Answer

Security, because the model was manipulated by malicious input

Question 4

A financial analyst is using a gen AI model to solve multi-step word problems involving budget allocations and percentage calculations. The model often produces the correct setup but arrives at the wrong final number, jumping directly to an answer without showing intermediate steps. Which prompt engineering technique is most likely to improve the accuracy of these calculations?

Accepted Answer

Chain-of-thought prompting, asking the model to reason through the problem step by step before giving the final answer

Answer

Zero-shot prompting, giving only the problem with no examples or reasoning guidance

Answer

Lowering the temperature setting so the model produces more deterministic output

Answer

Retrieval Augmented Generation to pull the correct answer from an external database

Question 5

A financial analyst uses a gen AI model to solve multi-step budget allocation word problems. When she asks the model to give the final answer directly, it frequently makes arithmetic and logic errors. She wants to improve accuracy without collecting labeled examples or tuning the model. Which prompt engineering technique should she apply?

Accepted Answer

Add an instruction asking the model to work through the problem step by step before giving the final answer

Answer

Provide five fully worked example problems with their answers in the prompt

Answer

Lower the temperature setting to zero so the model is more deterministic

Answer

Fine-tune the model on a dataset of solved budget problems

Question 6

A retail company deploys a customer-service chatbot built on a large language model. During testing, an employee asks about the return policy, and the chatbot confidently states that customers have 90 days to return items — but the actual company policy is 30 days, and this number appears nowhere in the model's training data or prompt. Which gen AI concept does this behavior best illustrate?

Accepted Answer

Hallucination — the model generated fluent but factually incorrect information not grounded in any source

Answer

A context window limitation that truncated the policy document

Answer

Embedding drift caused by outdated vector representations

Answer

Prompt injection introduced by the testing employee

Question 7

A marketing team uses a gen AI model to generate product descriptions. They have written three different prompt versions and want to systematically determine which one produces the highest-quality descriptions before rolling one out to production. What is the most effective way to evaluate which prompt performs best?

Accepted Answer

Run each prompt version against the same set of representative test inputs and score the outputs against a defined quality rubric, then compare the aggregate results

Answer

Choose the prompt with the most detailed instructions, since longer prompts always yield better output

Answer

Deploy all three prompts to production simultaneously and let customers decide which descriptions they prefer

Answer

Pick the prompt that produced the best result on a single sample product and assume it will generalize

Question 8

A product team is building a customer-support assistant on a large language model. During long conversations, the assistant starts to 'forget' details the customer mentioned near the beginning of the chat. An engineer explains this is due to a limit on how much text the model can consider at once. Which core gen AI concept is the engineer describing?

Accepted Answer

The context window — the maximum amount of input and output text (measured in tokens) the model can process in a single interaction

Answer

An embedding — a numeric vector representation of text used to measure semantic similarity

Answer

A hallucination — content the model generates that is fluent but factually incorrect

Answer

A prompt template — a reusable structure for formatting instructions sent to the model

Question 9

A product analyst pastes a very long transcript into a gen AI assistant and asks for a summary. The tool responds but appears to have ignored the earliest parts of the transcript, summarizing only the later sections. What is the most likely cause?

Accepted Answer

The input exceeded the model's context window, so the earliest content was pushed out and not processed.

Answer

The model hallucinated because the transcript contained factual errors.

Answer

The embeddings for the early text were corrupted during upload.

Answer

The model applied bias mitigation that removed the opening sections.

Question 10

A financial services firm wants to fine-tune a gen AI assistant on its internal documents, which include a mix of public marketing material, confidential client records, and regulated PII. Before any data is used for training, the Chief Data Officer wants to ensure the firm avoids inadvertently exposing sensitive information through the model. Which governance step should be completed FIRST?

Accepted Answer

Classify and label all source data by sensitivity level so that regulated and confidential records can be excluded or handled with appropriate controls

Answer

Deploy the fine-tuned model to a small pilot group and monitor its outputs for leaked information

Answer

Increase the model's temperature setting to reduce the chance it reproduces training data verbatim

Answer

Publish an AI transparency notice informing users that the assistant was trained on internal documents

Question 11

A retail company wants to build a gen AI assistant that answers employee questions using data from its HR system, finance platform, and internal wiki. During discovery, the project lead finds that the three systems use inconsistent formats, contain many duplicate and outdated records, and are managed by separate teams with no shared access policy. Leadership is eager to launch quickly. What should the project lead prioritize FIRST to set the initiative up for success?

Accepted Answer

Establish a data governance framework that addresses data quality, consistent access policies, and ownership across the three systems before large-scale development

Answer

Immediately fine-tune a foundation model on the combined raw data to meet the leadership timeline

Answer

Select the largest available model so its capacity can compensate for the inconsistent and duplicate data

Answer

Launch the assistant on the internal wiki only and ignore the HR and finance systems permanently

Question 12

A marketing analyst writes a prompt that includes both instructions ('Summarize the following customer review in one sentence') and the actual review text, all run together in a single paragraph. The model sometimes treats parts of the review as new instructions, producing inconsistent summaries. Without changing models or adding training data, what is the simplest prompt engineering improvement to make the output more reliable?

Accepted Answer

Use clear delimiters (such as triple quotes or XML-style tags) to separate the instruction from the review text the model should act on

Answer

Fine-tune the model on thousands of labeled review summaries

Answer

Raise the temperature setting so the model generates more varied summaries

Answer

Add a retrieval step that pulls related reviews from a vector database

Question 13

A product team is explaining to executives how their new gen AI feature can group customer feedback by meaning rather than exact keywords. They want to describe the underlying concept that lets the system understand that 'the app keeps crashing' and 'it freezes constantly' are semantically similar. Which core generative AI concept are they describing?

Accepted Answer

Embeddings, which represent text as numeric vectors so that similar meanings are positioned close together in a shared space

Answer

Tokens, which split text into smaller units that the model processes one at a time

Answer

Context windows, which define how much text the model can consider at once

Answer

Hallucinations, which occur when the model produces plausible but incorrect output

Question 14

A financial services company wants to build an internal tool that lets employees search a large knowledge base using natural language. When someone asks 'How do I reset a locked customer account?', the tool should return documents that are conceptually related even if they use different wording, such as 'unlock a client profile.' Which generative AI concept most directly enables this semantic matching between the query and the documents?

Accepted Answer

Embeddings that represent text as numerical vectors capturing semantic meaning

Answer

A larger context window that lets the model process more tokens at once

Answer

Prompt engineering that rephrases the user's question multiple times

Answer

Reducing hallucinations by lowering the model's temperature setting

Question 15

A product team has deployed a gen AI assistant that drafts customer email responses. Leadership wants a reliable way to measure whether output quality is actually improving as the team iterates on prompts and grounding. Automated metrics alone haven't correlated well with what customers consider a 'good' reply. What is the most effective approach to evaluate and improve output quality in this situation?

Accepted Answer

Establish a structured evaluation rubric scored by human reviewers against defined criteria (accuracy, tone, completeness) and track scores across iterations

Answer

Increase the model's temperature setting so responses become more varied and natural-sounding

Answer

Rely solely on the number of tokens generated as a proxy for how thorough each reply is

Answer

Switch to the largest available foundation model, since bigger models always produce higher-quality output

Question 16

A product team at an insurance company has deployed a gen AI assistant that answers policy questions. They frequently adjust prompts to improve responses, but they worry that a change fixing one type of question might silently break answers to other questions. What practice should they adopt to reliably measure whether each prompt change improves overall output quality?

Accepted Answer

Maintain a curated evaluation dataset of representative questions with expected answers, and re-run it against the model after every prompt change to compare results

Answer

Increase the model's temperature setting so responses become more varied and cover more cases

Answer

Rely on individual team members spot-checking a few answers each time they change a prompt

Answer

Fine-tune the model on the full policy documents so prompt changes are no longer necessary

Question 17

A media company deploys a gen AI tool that summarizes news articles. Early users complain that some summaries are fluent and well-written but occasionally contradict details in the original article. The product team wants to measure this specific problem systematically before their next model update. Which evaluation approach most directly targets the issue they are seeing?

Accepted Answer

Score each summary for factual consistency (faithfulness) against its source article, using human raters or an automated grounding check

Answer

Measure the average response latency of the summarization tool across a large sample of articles

Answer

Count the number of tokens in each summary to confirm outputs stay within a fixed length budget

Answer

Survey users on how readable and grammatically polished they find the summaries

Question 18

A marketing team uses a general-purpose LLM to generate product taglines. When they simply ask 'Write a tagline for our eco-friendly water bottle,' the outputs are generic and inconsistent in tone. The team wants better results quickly, without collecting a large dataset, training a custom model, or connecting to any external data source. What is the most appropriate technique to improve output quality in this situation?

Accepted Answer

Provide the model with several example taglines that reflect the desired tone and style within the prompt (few-shot prompting)

Answer

Fine-tune the model on thousands of historical marketing taglines to permanently adjust its behavior

Answer

Implement Retrieval Augmented Generation to pull taglines from an external database at generation time

Answer

Lower the model's temperature setting to zero to force more deterministic wording

Question 19

A financial analyst uses a gen AI model to extract key risk factors from lengthy 10-K filings. With a simple zero-shot instruction, the model produces inconsistent output — sometimes bullet lists, sometimes paragraphs, and occasionally it invents risks not present in the document. The analyst wants to improve consistency and accuracy without the cost and time of collecting a large labeled dataset or standing up new infrastructure. What is the most appropriate FIRST technique to try?

Accepted Answer

Provide a few example filings paired with correctly formatted, document-grounded outputs directly in the prompt

Answer

Fine-tune a foundation model on thousands of manually annotated 10-K filings

Answer

Increase the model's temperature setting to encourage more varied responses

Answer

Deploy a larger, more expensive model variant to a dedicated endpoint

Question 20

A support operations manager wants a gen AI model to categorize incoming customer tickets into one of five predefined categories using a very specific label format (e.g., 'CAT-3: Billing'). Early tests with a plain instruction produce inconsistent formatting and occasional made-up categories. The manager cannot train or tune a model and needs a quick fix within the prompt itself. Which prompt engineering technique should they apply first?

Accepted Answer

Provide a few labeled example tickets with their correct category outputs directly in the prompt (few-shot prompting)

Answer

Fine-tune the foundation model on thousands of historical labeled tickets

Answer

Increase the model's temperature setting to encourage more varied category assignments

Answer

Ask the model to explain its reasoning step by step before answering (chain-of-thought)

Question 21

A marketing analyst wants a foundation model to generate product descriptions that follow the company's very specific style: a punchy one-line headline, exactly three bullet points, and a closing call-to-action. When they simply describe these rules in a plain instruction, the model's output is inconsistent — sometimes it skips the call-to-action or writes four bullets. The analyst does not have a labeled dataset and needs a quick fix without any training. Which prompt engineering technique should they apply first?

Accepted Answer

Include several complete example descriptions written in the exact desired format within the prompt (few-shot prompting)

Answer

Fine-tune the foundation model on hundreds of past product descriptions

Answer

Add a chain-of-thought instruction telling the model to reason step by step before answering

Answer

Connect the model to a retrieval system to ground responses in the product catalog

Question 22

A marketing team wants a gen AI model to rewrite product descriptions in their distinctive brand voice. They only have about 15 well-written examples of the desired style, no ML engineers, and need a working solution within a day for a low-volume campaign. Which approach best fits these constraints?

Accepted Answer

Include several of the examples directly in the prompt as few-shot demonstrations of the desired style

Answer

Fine-tune a foundation model on the 15 examples to permanently encode the brand voice

Answer

Build a RAG pipeline that retrieves the 15 examples from a vector database at query time

Answer

Train a new model from scratch using the examples as the training corpus

Question 23

A product team uses a foundation model to convert customer feedback into structured JSON with fields for sentiment, category, and priority. Using a simple instruction prompt, the model returns correct content but with inconsistent key names, varying capitalization, and occasionally missing fields. The team cannot change the model and wants a fast prompt-only fix that enforces a consistent output structure. What should they do?

Accepted Answer

Add a few example input-output pairs showing the exact desired JSON structure directly in the prompt

Answer

Fine-tune the model on thousands of labeled JSON examples to lock in the format

Answer

Increase the model's temperature so it explores more formatting variations

Answer

Set up a Retrieval Augmented Generation pipeline to pull formatting rules from a document store

Question 24

A financial services company needs its gen AI assistant to consistently produce loan summaries in a highly specific internal format and tone that the base model struggles to replicate, even after extensive prompt engineering with detailed examples in every prompt. They have thousands of high-quality example summaries written by their analysts. Which technique will most reliably bake this behavior into the model so it no longer needs lengthy examples in each prompt?

Accepted Answer

Fine-tuning the model on their labeled dataset of example summaries

Answer

Adding a Retrieval Augmented Generation (RAG) pipeline over the summaries

Answer

Lowering the temperature setting to make outputs more deterministic

Answer

Continuing to expand the number of few-shot examples in the prompt

Question 25

A healthcare company uses a foundation model to generate patient-facing message drafts. Despite detailed prompts and several examples, the model still fails to consistently adopt the specialized empathetic clinical tone and terminology the organization requires across thousands of daily requests. They have a large curated dataset of high-quality past messages written by their clinical staff. Which approach will most reliably embed this consistent style and vocabulary into the model's default behavior?

Accepted Answer

Fine-tune the model on the curated dataset of past clinical messages

Answer

Add more few-shot examples to every prompt at inference time

Answer

Implement Retrieval Augmented Generation over the past message archive

Answer

Lower the temperature setting to make outputs more deterministic

Question 26

A financial services company wants its gen AI assistant to consistently adopt a very specific writing style and reasoning pattern used by its senior analysts. The company has 5,000 high-quality example question-answer pairs written in that exact style. The underlying facts change rarely. Which technique is the BEST fit for teaching the model to reliably produce output in this specialized style?

Accepted Answer

Fine-tuning the model on the 5,000 example pairs

Answer

Retrieval Augmented Generation (RAG) connected to a document store

Answer

Zero-shot prompting with a detailed instruction

Answer

Increasing the model's temperature setting

Question 27

A retail company wants a single AI system that its teams can reuse for several distinct tasks: drafting product descriptions, answering customer emails, and generating internal training summaries. Leadership is deciding whether to build separate specialized models for each task or adopt one large model that can be adapted to all of them. Which characteristic of foundation models best supports choosing the single-model approach?

Accepted Answer

Foundation models are trained on broad, large-scale data and can be adapted to many downstream tasks through prompting or fine-tuning.

Answer

Foundation models guarantee factually accurate outputs, eliminating the need for human review across tasks.

Answer

Foundation models are purpose-built for one narrow task and must be retrained from scratch for each new use case.

Answer

Foundation models only process numerical data, making them ideal for structured retail inventory tasks.

Question 28

A software company wants to build several internal tools: a document summarizer, a code-comment generator, and a customer-email drafter. Their data science lead suggests starting from a single large pre-trained model rather than building three separate systems from scratch. What key property of foundation models makes this approach practical?

Accepted Answer

Foundation models are trained on broad data at scale and can be adapted to many different downstream tasks

Answer

Foundation models are trained on a specific labeled dataset for one narrow task and cannot be reused elsewhere

Answer

Foundation models eliminate the need for prompts because they infer intent automatically without any input

Answer

Foundation models guarantee factually accurate outputs, removing the need for human review across all tasks

Question 29

A product manager is explaining to executives why the company's new generative AI assistant, built on a foundation model, can handle a wide range of tasks it was never explicitly programmed for. Which characteristic of foundation models best explains this broad capability?

Accepted Answer

They are trained on massive, broad datasets in a self-supervised way, allowing them to generalize across many downstream tasks.

Answer

They are hand-coded with explicit rules for each individual business task they need to perform.

Answer

They are trained on a small, narrowly labeled dataset specific to a single use case.

Answer

They store every possible answer in a lookup database and retrieve exact matches at runtime.

Question 30

A marketing director wants her team to use the consumer Gemini app to brainstorm campaign ideas, but the legal department is concerned that any confidential product details typed into prompts could be used to train Google's models. Which Google offering directly addresses this concern by providing enterprise-grade data governance where prompts and outputs are not used to train the models?

Accepted Answer

Gemini for Google Workspace (Gemini Enterprise) under the organization's Workspace account

Answer

The free public Gemini app accessed with personal Google accounts

Answer

A public LLM demo hosted on a third-party website

Answer

Copying prompts into a personal note-taking app before submitting them

Question 31

A small marketing team wants a quick, browser-based way to upload a product photo, ask questions about it, and generate promotional taglines in a chat interface. They have no engineers and no cloud project set up, and they don't need to integrate the tool into any application. Which Google offering best fits their needs?

Accepted Answer

The Gemini app, which provides a ready-to-use conversational interface supporting text and image inputs

Answer

Vertex AI Model Garden, to browse and deploy a foundation model to an endpoint

Answer

Vertex AI Agent Builder, to create a custom conversational agent grounded in a data store

Answer

The pre-built Vision API, to programmatically analyze the uploaded image

Question 32

A marketing manager at a mid-sized company wants a simple, chat-based tool that individual team members can use daily to brainstorm campaign ideas, draft social posts, and ask general questions—without any coding, infrastructure setup, or developer involvement. Which Google offering best fits this need?

Accepted Answer

The Gemini app, accessed through a web browser or mobile app

Answer

Vertex AI Model Garden, to deploy a foundation model on a managed endpoint

Answer

Vertex AI Agent Builder, to construct a conversational agent

Answer

The Vertex AI tuning pipeline, to fine-tune a custom model on marketing data

Question 33

A software engineering team at a retail company wants an AI assistant that is integrated directly into their IDE to help developers write, explain, and refactor code more quickly, while keeping their proprietary source code protected. Which Google Cloud offering is the most appropriate fit?

Accepted Answer

Gemini Code Assist

Answer

The consumer Gemini app on gemini.google.com

Answer

Vertex AI Model Garden

Answer

Cloud Natural Language API

Question 34

A retail company wants to add an AI feature that generates short product description snippets for millions of catalog items each day. The engineering lead is concerned about keeping per-request latency and cost low, since the task is relatively simple and does not require the most advanced reasoning. Which approach best fits this requirement when selecting a Gemini model?

Accepted Answer

Choose a smaller, faster Gemini model variant (such as a Flash-tier model) optimized for high-volume, low-latency, cost-sensitive tasks

Answer

Always use the largest, most capable Gemini model variant to guarantee the highest possible output quality regardless of cost

Answer

Use a predictive machine learning classifier since generating text at scale is not a generative AI use case

Answer

Avoid Gemini entirely and build a custom model from scratch on TPUs to minimize inference cost

Question 35

A publishing company wants to feed an entire 800-page novel into a single request so a gen AI model can analyze plot consistency, track character arcs across all chapters, and answer questions referencing any part of the book at once. Which characteristic of Google's Gemini models makes this possible without splitting the book into many separate requests?

Accepted Answer

Gemini's large context window, which allows very large amounts of text to be processed together in a single prompt

Answer

Gemini's low temperature setting, which ensures deterministic and repeatable answers

Answer

Gemini's fine-tuning capability, which permanently stores the book in the model's weights

Answer

Gemini's multimodal capability, which lets it accept images alongside text

Question 36

A retail company already uses Google Workspace across all departments. The sales team wants an assistant embedded directly inside Gmail and Google Docs that can summarize customer email threads and draft replies using the content in those messages. They do not want to build or host any custom model, and they have no engineering resources to write integration code. Which Google Cloud offering best fits this need?

Accepted Answer

Gemini for Google Workspace

Answer

Vertex AI Studio for prompt prototyping

Answer

A tuned Gemini model deployed on a Vertex AI endpoint

Answer

Vertex AI Agent Builder with a custom data store

Question 37

A retail company wants to build a customer-facing mobile app that uses a Gemini model to answer product questions, and they need programmatic API access, the ability to tune the model on their catalog, and control over deployment. A leader suggests they simply buy Gemini for Google Workspace licenses to accomplish this. Why is this recommendation inappropriate for their goal?

Accepted Answer

Gemini for Google Workspace enhances productivity apps like Gmail and Docs for internal users, not building custom applications; Vertex AI is the platform for programmatic access, tuning, and deployment.

Answer

Gemini for Google Workspace cannot access Gemini models at all, so it would return no useful responses to product questions.

Answer

Gemini for Google Workspace is more expensive per user than Vertex AI, making it the wrong financial choice for this project.

Answer

Gemini for Google Workspace only supports text summarization and cannot answer any customer questions.

Question 38

A marketing manager at a mid-sized company has already written a rough draft of a customer newsletter in Google Docs. She wants an AI assistant that can rewrite the draft in a friendlier, more casual tone directly within the document, without exporting the text elsewhere or building any custom application. Which Google offering is the most appropriate?

Accepted Answer

Gemini for Google Workspace, using the assistive writing features inside Google Docs

Answer

Vertex AI Studio, to prototype prompts that rewrite the newsletter

Answer

Vertex AI Agent Builder, to create a conversational agent that edits documents

Answer

The pre-built Natural Language API, to analyze and adjust the tone of the text

Question 39

A regional sales manager spends hours each morning writing individualized email replies to prospects directly within Gmail. She wants an AI feature that can draft context-aware responses inside her existing inbox, using the content of the email thread she is reading, without switching to a separate application or building anything custom. Which Google offering best fits this need?

Accepted Answer

Gemini for Google Workspace, which brings AI assistance directly into Gmail to draft and refine replies

Answer

Vertex AI Agent Builder to create a conversational agent for email handling

Answer

The pre-built Natural Language API to analyze the sentiment of incoming emails

Answer

Vertex AI Model Garden to select and tune an open model for email generation

Question 40

A global sales team holds video meetings where participants speak English, Spanish, and Japanese. The manager wants attendees to see live translated captions during Google Meet calls without deploying any custom infrastructure or writing code. Which Google offering best meets this need?

Accepted Answer

Gemini for Google Workspace, which provides translated captions natively within Google Meet

Answer

A Vertex AI custom model trained on the team's multilingual meeting transcripts

Answer

The Cloud Translation API integrated through a custom-built middleware service

Answer

Vertex AI Agent Builder configured with a conversational translation agent

Question 41

A project manager runs weekly video calls with her cross-functional team using Google Meet. She spends significant time after each meeting writing up notes and tracking who agreed to do what. She wants an integrated capability that automatically captures the discussion and produces a summary with action items, without adopting a separate third-party tool. Which Google offering best meets this need?

Accepted Answer

Gemini for Google Workspace, using its 'take notes for me' capability in Google Meet

Answer

Vertex AI Agent Builder to create a custom conversational agent for meetings

Answer

The pre-built Speech-to-Text API to transcribe the audio recording manually afterward

Answer

Vertex AI Model Garden to select and deploy an open summarization model

Question 42

A retail company's IT director wants employees to use generative AI directly inside their existing productivity tools—drafting emails in Gmail, summarizing documents, and generating content in Docs—without requiring the team to build or manage any AI infrastructure. The director also insists that no company data be used to train Google's foundation models. Which Google Cloud offering best fits these requirements?

Accepted Answer

Gemini for Google Workspace, using the enterprise edition with data governance protections

Answer

Vertex AI with a tuned Gemini model deployed to a custom endpoint

Answer

The consumer Gemini app accessed through personal Google accounts

Answer

Vertex AI Agent Builder connected to the company's document data store

Question 43

A finance analyst at a retail company spends hours each week in Google Sheets building formulas and categorizing sales data. She wants an AI assistant that works directly inside her existing spreadsheets to help generate formulas, classify entries, and organize data without leaving the tool. Which Google offering best fits this need?

Accepted Answer

Gemini for Google Workspace, which brings AI assistance directly into apps like Sheets, Docs, and Gmail

Answer

Vertex AI Model Garden, to select and deploy an open model for spreadsheet processing

Answer

The pre-built Natural Language API, to analyze the text in her spreadsheet cells

Answer

Vertex AI Agent Builder, to create a conversational agent that answers finance questions

Question 44

A marketing manager at a retail company needs to quickly create a first-draft slide deck for an upcoming product launch. She already has the key talking points written in a Google Doc and wants to generate polished presentation slides directly within her existing Google Workspace tools, without switching to a separate application or writing any code. Which Google offering best fits this need?

Accepted Answer

Gemini for Google Workspace, using its generative features inside Google Slides

Answer

Vertex AI Agent Builder to create a conversational agent that produces slides

Answer

The pre-built Vision API to convert her document text into images

Answer

Vertex AI Model Garden to select and deploy an open image-generation model

Question 45

A marketing manager spends hours each morning reading long email threads in Gmail and reviewing lengthy documents in Google Docs. She wants an AI capability built directly into these tools to summarize threads, draft replies, and refine document text without copying content into a separate application. Which Google offering best meets this need?

Accepted Answer

Gemini for Google Workspace

Answer

Vertex AI Model Garden

Answer

Vertex AI Agent Builder

Answer

The consumer Gemini app on her phone

Question 46

A learning and development manager wants her team to quickly produce short internal training videos with narration and on-screen text, using existing company documents as source material. She wants a solution built into the Google Workspace tools her team already uses, without hiring video editors or learning a separate production platform. Which Google offering best fits this need?

Accepted Answer

Google Vids with Gemini for Google Workspace

Answer

Vertex AI Agent Builder to create a conversational training agent

Answer

The pre-built Video Intelligence API

Answer

A custom model tuned in Vertex AI Model Garden

Question 47

A financial analyst at a mid-sized firm spends hours each week manually building formulas and summarizing trends inside large spreadsheets stored in Google Sheets. Leadership wants to boost the analyst's productivity within their existing office tools without any custom development or new platform. Which Google offering best fits this need?

Accepted Answer

Gemini for Google Workspace

Answer

Vertex AI Agent Builder

Answer

Vertex AI Model Garden

Answer

The pre-built Natural Language API

Question 48

A logistics company wants employees to ask natural-language questions and get answers grounded in their internal policy documents, HR handbooks, and standard operating procedures stored across the organization. Leadership wants a searchable, enterprise-ready assistant that surfaces relevant internal content, with minimal custom development. Which Google Cloud offering is the most appropriate fit?

Accepted Answer

Google Agentspace, which provides an enterprise search and AI assistant grounded in the company's connected internal data sources

Answer

The consumer Gemini app, where each employee pastes documents into the chat before asking questions

Answer

Vertex AI custom training to build a foundation model from scratch on the company's documents

Answer

The pre-built Cloud Natural Language API to classify the sentiment of the internal documents

Question 49

An HR department wants to launch a customer-facing web portal where new job applicants can ask questions and receive answers grounded in the company's public careers FAQ and benefits documents. The portal must be embeddable on the company's external recruiting website, handle multi-turn conversations, and connect to an internal API that checks application status. Which Google Cloud offering is the best fit?

Accepted Answer

Vertex AI Agent Builder, to create a grounded conversational agent connected to the FAQ documents and the status-checking API

Answer

Gemini for Google Workspace, so HR staff can draft FAQ responses inside Google Docs

Answer

The Gemini app with an enterprise data protection plan for the HR team to use internally

Answer

The pre-built Natural Language API to analyze the sentiment of applicant questions

Question 50

A software company wants to embed a generative AI feature directly into their proprietary SaaS product so that their own customers can auto-generate summaries of support tickets stored in the company's application database. The development team needs programmatic access to a Gemini model, the ability to control prompts in their code, and options to scale and monitor the deployment. Which Google Cloud offering best fits this requirement?

Accepted Answer

Vertex AI, using the Gemini API to integrate the model into the application programmatically

Answer

The Gemini app (gemini.google.com) accessed by each customer through a browser

Answer

Gemini for Google Workspace enabled across the company's Gmail and Docs

Answer

Google Cloud's pre-built Natural Language API for entity extraction

Question 51

A retail company ran a successful gen AI pilot for product description generation using a premium model. Finance approved a budget based on the pilot's monthly costs. Six months after full rollout, actual costs are three times the projected budget, even though usage grew only modestly. The leadership team wants to understand the most likely root cause of the cost overrun so they can plan future initiatives more accurately. What is the MOST likely explanation?

Accepted Answer

The pilot budget only accounted for per-request inference costs and overlooked ongoing costs such as monitoring, human review, data pipeline maintenance, and prompt/output token growth at scale.

Answer

Premium foundation models automatically increase their per-token price once an organization exceeds a fixed number of monthly requests.

Answer

Gen AI models require full retraining every month, and retraining costs were excluded from the pilot budget.

Answer

The company should have used a predictive ML model instead, since generative models cannot be budgeted for at all.

Question 52

A European retail bank wants to deploy a gen AI assistant that processes customer transaction data. Regulators require that all customer data remains stored and processed within the EU, and the bank must be able to demonstrate this to auditors. As the AI leader evaluating deployment options, which consideration should most influence the platform and configuration you select?

Accepted Answer

Selecting a deployment region and data governance controls that enforce EU data residency and provide auditable processing logs

Answer

Choosing the foundation model with the largest context window to handle full transaction histories

Answer

Picking the cheapest model variant to minimize inference costs at scale

Answer

Prioritizing the model with the fastest response latency for a good customer experience

Question 53

A retail company launched six gen AI initiatives 18 months ago. During a portfolio review, an AI-powered product-description generator shows high monthly inference costs but a marginal, hard-to-attribute lift in conversion, while consuming significant engineering support time. Three other initiatives show strong, measurable ROI. As the gen AI program leader, what is the most appropriate governance action for the product-description generator?

Accepted Answer

Decommission or pause it after documenting the decision, reallocating budget and engineering capacity to the higher-ROI initiatives.

Answer

Immediately scale it to all product categories, since sunk investment justifies broader deployment.

Answer

Keep it running indefinitely without review, because removing a live AI feature would signal program failure to executives.

Answer

Switch it to a larger, more expensive foundation model to try to improve the conversion lift before evaluating cost.

Question 54

A retail company's leadership team is evaluating new technology and asks their innovation lead to explain what fundamentally distinguishes generative AI from the analytics tools they already use. Which description best captures the defining capability of generative AI?

Accepted Answer

It creates new, original content such as text, images, or code based on patterns learned from training data

Answer

It classifies existing records into predefined categories with high accuracy

Answer

It stores and retrieves structured business data faster than a traditional database

Answer

It automates repetitive rule-based tasks by following explicitly coded if-then logic

Question 55

A product manager at a media company is explaining to executives how the new generative AI writing assistant fundamentally works, compared to the company's older spam-detection system. Which statement best captures the defining behavior of the generative AI tool?

Accepted Answer

It learns statistical patterns from large amounts of text and produces new, original content in response to a prompt.

Answer

It assigns each incoming item to one of a fixed set of predefined categories based on labeled training examples.

Answer

It stores every document it has seen and retrieves the closest exact match when asked a question.

Answer

It calculates a single numeric probability score to predict whether a future event will occur.

Question 56

A retail company rolled out a gen AI product-description generator across its e-commerce team six months ago. Leadership now claims the tool 'drove a 12% increase in online sales' and wants to expand the investment. The analytics lead is skeptical because online sales also benefited from a new seasonal ad campaign and a website redesign during the same period. What should the leader recommend before attributing the sales lift to the gen AI tool and approving expansion?

Accepted Answer

Isolate the tool's impact using a controlled comparison (e.g., an A/B test or holdout group) so the sales lift can be causally attributed rather than correlated with concurrent initiatives

Answer

Approve the expansion immediately, since a 12% sales increase clearly exceeds any reasonable ROI threshold for the tool

Answer

Measure the number of product descriptions generated per week and report that as the primary business impact metric

Answer

Survey the e-commerce team about whether they feel the tool is helpful and use their sentiment as the ROI justification

Question 57

A retail company ran three successful gen AI pilots in isolated teams, each showing strong ROI. However, when leadership tried to scale these solutions enterprise-wide, adoption stalled: budgets were fragmented, teams duplicated effort on the same tooling, and no one owned enterprise-wide decisions on vendor selection or Responsible AI standards. As the gen AI program leader, what is the most effective organizational action to unblock scaling?

Accepted Answer

Establish executive sponsorship and a centralized operating model (such as a Center of Excellence) to coordinate funding, standards, and shared tooling across teams

Answer

Ask each team to independently expand its own pilot and negotiate separate vendor contracts to preserve their autonomy and speed

Answer

Pause all gen AI work until every employee completes prompt-engineering training to close the skills gap

Answer

Select the single highest-ROI pilot and cancel the other two to focus all resources on one initiative

Question 58

A large financial services company is scaling multiple gen AI projects across departments. Different teams are independently deploying models, choosing their own vendors, and setting their own acceptable-use rules. Leadership is concerned about inconsistent risk handling, duplicated spend, and potential regulatory exposure. What is the MOST effective organizational-level action to address these concerns?

Accepted Answer

Establish a cross-functional AI governance body that sets enterprise-wide policies, approval processes, and risk standards for all gen AI initiatives

Answer

Require every team to fine-tune its own model so outputs are tailored to each department's specific needs

Answer

Allow each department to continue operating independently but ask them to submit quarterly usage reports for review

Answer

Consolidate all gen AI work under a single engineering team that builds every solution to guarantee technical consistency

Question 59

A retail company ran a 3-month gen AI pilot for an internal customer-service assistant. Leadership now wants to decide whether to scale it enterprise-wide. The team has anecdotal praise from agents but no structured evidence. Before approving further investment, what should leadership require to make a defensible scaling decision?

Accepted Answer

Defined baseline and post-pilot KPIs (e.g., average handle time, first-contact resolution, cost per ticket) tied to the pilot's business objective

Answer

A larger foundation model with a longer context window to improve the assistant's answer quality

Answer

A count of how many prompts agents submitted during the pilot period

Answer

A survey asking agents whether they personally enjoyed using the tool

Question 60

A retail company's newly formed AI steering committee has approved four gen AI initiatives but has limited engineering capacity to launch them one at a time over the next year. The CFO wants to build organizational momentum and secure continued funding, while the CEO wants to ensure the program eventually delivers transformational value. Which sequencing approach best balances these leadership priorities when deciding what to build first?

Accepted Answer

Start with a low-effort, high-visibility use case that delivers a measurable early win, then reinvest the credibility and learnings into the more complex, higher-value initiatives

Answer

Launch the most technically ambitious and highest-value initiative first, since it will justify the entire program's budget in a single deliverable

Answer

Sequence the initiatives strictly by which teams volunteered first, to reward enthusiasm and maximize internal buy-in

Answer

Delay all launches until every initiative has a fully validated ROI model, so the committee can rank them purely by projected financial return

Question 61

A retail company launched a gen AI shopping assistant six months ago. In the quarterly review, the product team proudly reports that the assistant handled 2 million conversations and generated over 500,000 product recommendations. The CFO is unconvinced and asks the Gen AI Leader to demonstrate the solution's actual business value. Which metric should the leader prioritize to answer the CFO's question?

Accepted Answer

The incremental revenue and conversion rate lift attributable to sessions using the assistant, compared against a control group

Answer

The total number of conversations and recommendations the assistant produced

Answer

The average response latency and uptime of the assistant's underlying model endpoint

Answer

The number of employees trained to maintain and update the assistant

Question 62

A large insurer is launching several gen AI initiatives at once: an internal tool that summarizes meeting notes, a customer-facing chatbot that answers policy coverage questions, and a marketing team's brainstorming assistant. The Responsible AI governance board has limited review capacity and must decide where to focus its most rigorous oversight first. Which approach best reflects a risk-based governance strategy?

Accepted Answer

Prioritize the deepest review on the customer-facing policy chatbot, because incorrect coverage answers carry the highest regulatory and financial harm

Answer

Apply the same intensive review process equally to all three initiatives to ensure consistency

Answer

Prioritize the internal meeting-notes tool first, since it is deployed soonest and touches the most employees

Answer

Prioritize the marketing brainstorming assistant, because creative outputs are hardest to predict

Question 63

A retail company deployed a gen AI tool that drafts product descriptions. Six months in, the executive team wants proof of business impact before expanding it company-wide. The project lead's report highlights that 4,000 descriptions were generated and that 85% of the marketing team logs in weekly. The CFO says these figures do not answer whether the investment is paying off. Which additional metric would best demonstrate the tool's ROI?

Accepted Answer

The net financial benefit—reduced content-production hours and any lift in product-page conversions—compared against the tool's total cost

Answer

The number of prompt templates the marketing team has created since launch

Answer

The average length in tokens of each generated product description

Answer

The percentage of employees who rated the tool 'easy to use' in a satisfaction survey

Question 64

A retail company's leadership wants to launch its first gen AI initiative. Four proposals are on the table, each with different expected impact and feasibility. Leadership wants to pick the use case that best demonstrates measurable business value while managing risk for an initial pilot. Which proposal should they prioritize?

Accepted Answer

Automating responses to high-volume, repetitive customer service inquiries, where deflection rate and cost-per-ticket savings can be tracked against current metrics

Answer

Generating fully autonomous pricing decisions for all products without human oversight, promising large but unquantified margin gains

Answer

Building a company-wide 'AI everything' platform that touches every department simultaneously to maximize organizational exposure

Answer

Creating an experimental art-generation tool for the marketing team with no defined success metric, chosen because it is technically interesting

Exam domain	Exam weight	Practice
Fundamentals of gen AI	30%	Practice this topic
Google Cloud's gen AI offerings	35%	Practice this topic
Techniques to improve gen AI model output	20%	Practice this topic
Business strategies for a successful gen AI solution	15%	Practice this topic

Generative AI Leader (Gen AI Leader) practice exam & study guide

Start studying Gen AI Leader

All Gen AI Leader study resources

Gen AI Leader exam domains

Sample Gen AI Leader questions

Key Gen AI Leader terms

Gen AI Leader frequently asked questions

Fundamentals of gen AI (30%)

Google Cloud's gen AI offerings (35%)

Techniques to improve gen AI model output (20%)

Business strategies for a successful gen AI solution (15%)