Generative AI Leader · Free Practice Question Medium
Question 9
A news organization wants to build a generative AI tool that allows journalists to ask questions about current events and receive answers that are not only informative but also explicitly cite sources from reputable, up-to-date news websites on the public internet.
Which Google Cloud grounding offering would be most suitable for connecting their AI to this broad, publicly available "world data"?
-
A
RAG APIs with a custom vector database of historical archives.
-
B
Grounding with Google Search.
-
C
Fine-tuning a model exclusively on a curated dataset of books.
-
D
Pre-built RAG with Vertex AI Search using only internal company documents.
Reveal correct answer
Correct answer: B
Explanation
Grounding with Google Search allows generative AI applications to leverage the vast, up-to-date information indexed by Google's public search engine.
INCORRECT: Pre-built RAG with Vertex AI Search using only internal company documents.
This would ground the AI in internal data, not the public internet news sources required.INCORRECT: RAG APIs with a custom vector database of historical archives.
This would ground the AI in specific historical archives, not broadly across current, reputable news websites.CORRECT: Grounding with Google Search.
Google Cloud offers capabilities to ground generative AI models with Google Search. This allows the AI to access and cite information from the vast index of the public web, including up-to-date news websites, to provide well-sourced answers about current events.INCORRECT: Fine-tuning a model exclusively on a curated dataset of books.
This would limit the model's knowledge to the content of those books and wouldn't provide access to real-time, current news from web sources.
Discussion
Think the marked answer is wrong, or have a better explanation? Share it below — comments appear after review.
