Generative AI Leader Practice Test 32 — 15 Questions

Question 1

During a proof-of-concept, a team notices that their GenAI-powered meeting summarizer occasionally includes hallucinated details. They want to improve summary accuracy before production. Which action would be most effective?

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

Accepted Answer

Provide the meeting transcript as context in the prompt. Providing the meeting transcript as context grounds the model in factual data, reducing hallucinations. Few-shot prompting may help, but the primary issue is lack of source material. Adjusting temperature is less impactful. Switching to a larger model without context may not help.

Answer

Upgrade to a larger model

Answer

Decrease the temperature to 0.0

Answer

Use few-shot prompting with example summaries

Question 2

An enterprise wants to use Gemini for a customer-facing application. They require the following: data isolation in a VPC, audit logging, and SLA guarantees. Which THREE features of Vertex AI satisfy these requirements?

Accepted Answer

Vertex AI SLA (Service Level Agreement). Vertex AI offers a defined Service Level Agreement (SLA) that guarantees uptime and performance metrics for enterprise customers, which is a core requirement for customer-facing applications. The SLA provides contractual assurances, typically covering availability and response times, ensuring the enterprise can meet its own service commitments.

Answer

Gemini Nano on-device

Answer

Google AI Studio free tier

Question 3

A company wants to build a chatbot that can answer questions about recent news events up to the current date. They want to minimize hallucinations and ensure factual accuracy. Which Google Cloud feature should they integrate?

Accepted Answer

Gemini API with Google Search grounding. The Gemini API with Google Search grounding enables the chatbot to retrieve and cite real-time information from the web, directly addressing the need for up-to-date news and factual accuracy. Grounding connects the model's responses to verified search results, reducing hallucinations by anchoring answers in current, authoritative sources rather than relying solely on the model's training data.

Answer

Natural Language AI entity analysis

Answer

Document AI

Answer

Vertex AI Vector Search

Question 4

A startup wants to build a text-to-speech application for generating audiobooks. Which Google Cloud generative AI service is best suited for this task?

Accepted Answer

Chirp. Chirp is Google Cloud's text-to-speech AI model. Imagen generates images, Codey generates code, and Gemini is a multimodal model not specialized for high-quality speech synthesis.

Answer

Imagen

Answer

Codey

Answer

Gemini

Question 5

A machine learning engineer is deploying a large generative model on Vertex AI. The model requires a GPU with high memory. Which machine configuration should they choose?

Accepted Answer

a2-highgpu-4g with 4 A100 GPUs. The a2-highgpu-4g machine series is specifically designed for large-scale GPU-accelerated workloads, offering 4 NVIDIA A100 GPUs with 40GB of high-bandwidth memory (HBM2e) each, totaling 160GB of GPU memory. This configuration provides the high memory capacity required for training or serving large generative models, such as LLMs or diffusion models, which often exceed the memory limits of smaller GPUs.

Answer

c2-standard-16 with no GPU

Answer

n1-standard-4 with a single T4 GPU

Answer

n2-standard-8 with a single P4 GPU

Question 6

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?

Accepted Answer

Apply reinforcement learning from human feedback (RLHF). 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.

Answer

Increase the amount of training data

Answer

Lower the top_k sampling value

Answer

Increase the temperature parameter

Question 7

A healthcare organization wants to use generative AI for clinical decision support. They must comply with regulatory requirements for high-risk AI systems. Which THREE measures should they implement? (Choose 3)

Accepted Answer

Implement a human-in-the-loop review for all AI-generated recommendations. Option A is correct because high-risk clinical decision support requires human oversight so that a qualified clinician reviews and can override every AI-generated recommendation before it affects patient care. Option B is correct because transparency obligations for high-risk AI demand clear documentation of intended purpose, performance metrics, and known limitations so that deployers and regulators can assess fitness and residual risk. Option C is correct because regulatory frameworks for high-risk AI mandate ongoing risk management, including periodic assessments to identify, evaluate, and mitigate potential harms throughout the system lifecycle. Option D is not required, since fine-tuning on proprietary hospital data is a design choice and does not by itself satisfy regulatory compliance, and could even raise privacy concerns. Option E is not required, because on-premises deployment is one possible privacy control but is not mandated; compliant cloud deployment with appropriate safeguards can meet the same regulatory requirements.

Answer

Use a model that has been fine-tuned on proprietary hospital data

Answer

Deploy the system on-premises only to ensure data privacy

Question 8

A startup is building a generative AI tool that helps users write code. They want to launch quickly but need to ensure the generated code is secure and does not introduce vulnerabilities. They have a small team of developers with some ML experience. The tool should be cloud-hosted. Which approach balances speed, security, and cost?

Accepted Answer

Use a pre-trained code model (e.g., Codey) and add a security filtering layer. Using a pre-trained code model such as Codey provides immediate capability without the cost and time of training from scratch, while adding a security filtering layer (SAST-style scanning, output sanitization, policy checks) addresses the security requirement. This balances speed, security, and cost for a small team.

Answer

Deploy the tool without any security checks and rely on manual review

Answer

Train a custom code generation model from scratch on a large dataset

Answer

Use a smaller model and restrict outputs to only simple code patterns

Question 9

A developer is using Vertex AI Gemini API for a chatbot. The chatbot sometimes outputs harmful content. What is the best first step to mitigate this?

Accepted Answer

Use safety filters and safety settings in the API request. The Vertex AI Gemini API provides built-in safety filters and configurable safety settings (e.g., `safety_settings` parameter with categories like `HARM_CATEGORY_HARASSMENT` and thresholds like `BLOCK_ONLY_HIGH`) that allow developers to block harmful outputs at inference time without retraining. This is the fastest and most direct first step to mitigate harmful content, as it requires no additional infrastructure or model modification.

Answer

Fine-tune the model on curated safe data

Answer

Add a human-in-the-loop review

Answer

Switch to a smaller model

Question 10

A financial services firm is deploying a Gemini-based application that must comply with GDPR. The application processes customer queries and may include personal data. Which Google Cloud capability should the firm use to ensure that the model does not expose personally identifiable information (PII) in its responses?

Accepted Answer

Integrate Cloud Sensitive Data Protection (DLP) to inspect and redact PII from prompts and responses. Cloud Sensitive Data Protection (DLP) is designed to inspect, classify, and redact sensitive data such as PII. Integrating it with the Gemini application allows real-time scanning and redaction of PII from both prompts and responses, ensuring GDPR compliance by preventing exposure of personal data.

Answer

Set the model's safety filters to block all categories at maximum threshold

Answer

Use Vertex AI's Model Garden with a restricted access policy

Answer

Fine-tune the model on a dataset that has no PII

Question 11

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?

Accepted Answer

Use Google's Generative AI Studio with pre-trained models via API. 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).

Answer

Hire a team of data scientists to fine-tune an open-source model

Answer

Use a third-party platform that requires on-premise deployment

Answer

Build a custom foundation model from scratch using their client data

Question 12

A researcher wants to detect whether text was generated by an AI model to identify potential misinformation. Which Google technology is specifically designed for this purpose?

Accepted Answer

SynthID. SynthID is a Google DeepMind technology specifically designed to watermark and detect AI-generated text, images, audio, and video. It embeds an imperceptible digital watermark directly into the output of generative models, allowing for reliable detection even after modifications like cropping or compression. This makes it the correct tool for identifying potential misinformation from AI-generated text.

Answer

Model Cards

Answer

PAIR Explorables

Answer

Datasheets for Datasets

Question 13

A sales team needs to draft personalized emails to prospects. They want a tool integrated into Gmail that suggests completions and can generate full drafts based on brief prompts. Which Gemini for Google Workspace feature should they use?

Accepted Answer

Gemini in Gmail with Smart Compose and 'Help me write'. Smart Compose and 'Help me write' in Gmail provide AI-powered suggestions and draft generation directly within the email interface.

Answer

Gemini in Sheets with formula assistance

Answer

Gemini in Meet for meeting summaries

Answer

Gemini in Docs with 'Help me write'

Question 14

A company wants to adopt GenAI for code generation and review. To ensure code quality and security, they plan to implement a change management program. Which THREE actions are most effective?

Accepted Answer

Create a code review checklist specifically for AI-generated code. Option A is correct because a dedicated code review checklist for AI-generated code ensures reviewers systematically check for common LLM pitfalls such as hallucinated APIs, insecure patterns, license issues, and subtle logic errors, directly supporting code quality and security. Option B is correct because training on prompt engineering helps developers craft precise prompts, provide context, and constrain outputs, which improves the relevance, correctness, and security of generated code and reduces rework. Option E is correct because a limited pilot with a small developer group allows the organization to validate tooling, measure quality and security impacts, refine review processes, and gather feedback before a company-wide rollout, lowering risk. Option C is not appropriate because concentrating all responsibility in a single AI champion creates a bottleneck and a single point of failure rather than a scalable change management program. Option D is incorrect because phasing out senior developers removes the human expertise needed to review, validate, and secure AI-generated code, which would undermine quality and security rather than improve them.

Answer

Appoint a single AI champion to manage all code generation

Answer

Phase out senior developers to rely entirely on AI-generated code

Question 15

A product manager wants to communicate the limitations of a new generative AI feature to stakeholders. According to Google's People + AI Guidebook, what is the BEST approach?

Accepted Answer

Provide clear examples of the AI's capabilities and failure modes. The People + AI Guidebook emphasizes setting appropriate expectations by clearly communicating what the AI can and cannot do.

Answer

Highlight only the successful use cases to build excitement

Answer

Promise that future versions will overcome all limitations

Answer

Share a technical paper detailing the model architecture