GCP Generative AI Leader Practice Question 68: Fine Tuning vs Prompting

Google Cloud Generative AI Leader · Test 7 · question 8 of 10

Techniques to Improve Model Outputmedium

A legal team has spent months writing detailed prompt templates and few-shot examples to get a general-purpose model to draft contracts in the firm's exact house style and terminology. Despite extensive prompt tuning, the model still drifts from the required style and vocabulary on longer documents, and the firm has thousands of past contracts labeled by clause type available to use as training data.

What should the team do next?

Select one answer
Show answer and explanation

Answer:

  • A. Fine-tune the model on the firm's labeled contract examples, since prompting alone has plateaued and a large, high-quality labeled dataset is available to adjust the model's parameters

Fine-tuning is the right move exactly when prompt engineering has been pushed as far as it can go and a sizable, high-quality labeled dataset exists: it adjusts the model's parameters directly so the desired style and terminology become baked into the model itself, rather than being re-explained in every prompt. The scenario describes both conditions, months of prompt tuning that still fall short, and thousands of labeled examples ready to use.

- B is wrong because the team has already invested heavily in prompt engineering and it has plateaued; repeating the same style guide more verbosely is unlikely to fix persistent drift on long documents. - C is wrong because swapping models without using the firm's available labeled data throws away the asset best suited to solving the exact problem, consistent house style. - D is wrong because raising temperature increases output randomness, which works against the goal of consistent, predictable house style rather than helping it.

Reference: About supervised fine-tuning for Gemini models

Practice Test 7 ›

Study guides for Google Cloud Generative AI Leader

Free AI certifications in 2026: what's really $0Which free AI certifications are real exams, which are badges, and which are course certificates: Oracle, Microsoft, Google, AWS, Claude and more.
Google Cloud roadmap: GCP zero to hero in 10 stagesA Google Cloud roadmap from zero to hero: 10 stages, 84 topics in learning order, free-tier labs, exam traps and which certification fits each stage.
Google Cloud Digital Leader exam: what's on it, who it's forGoogle Cloud Digital Leader: August 2026 exam update, six sections, US$99 price as of November 2026, renewal, study plan and ACE comparison.
Agentic AI certification: Oracle, Google, GitHub, NVIDIAAgentic AI certifications in 2026: Oracle's free exam, GitHub GH-600, Google, NVIDIA and Claude, compared by status, price and role.
Google Cloud Generative AI Leader vs AWS AI PractitionerGoogle Cloud Generative AI Leader vs AWS AI Practitioner: format, cost, topics, free preparation and which certification to take first.
Best AI certifications 2026: AWS, Google, Microsoft, ClaudeThe best AI certifications in 2026 from AWS, Google, Microsoft, Claude and Databricks, by level, with price, status and which one to take first.
Google Cloud certification path: is ACE worth it?Is Google's Associate Cloud Engineer certification worth it? The Google Cloud certification path in 2026, with cost, format and which exam to take first.
Google Professional Agentic Architect: what it isGoogle's Professional Agentic Architect certification explained - beta status, the two-part exam, the five domains, price, validity and how to prepare.
What Is RAG? How Retrieval-Augmented Generation WorksWhat RAG is and how it works: chunking, embeddings, vector search, hybrid search, reranking, GraphRAG and agentic RAG on AWS, Google and Azure.
Google Cloud's generative AI offerings, mappedGemini app, Gemini for Workspace, Gemini Enterprise, Agent Search, the contact center suite and Agent Platform: which one a business should pick, and why.
Prompt engineering vs RAG vs fine-tuningWhen to use prompt engineering, retrieval-augmented generation, fine-tuning or continued pre-training, compared on cost, freshness, hallucination and privacy.