GCP Generative AI Leader Practice Question 19: Knowledge Cutoff Public Information
Google Cloud Generative AI Leader · Test 2 · question 9 of 10
A travel publisher's assistant drafts destination briefings for corporate clients. It is accurate on history and geography, but it states visa rules and airline route changes that were superseded months ago, and it has no idea about a border crossing that reopened last week. None of this information is the publisher's own; it is public, it changes constantly, and the briefings must reflect today's position with sources clients can follow.
Which approach addresses the limitation the assistant is running into?
Show answer and explanation
Answer:
The assistant is hitting the model's knowledge cutoff: a foundation model only knows what was in its training data, so anything that changed afterwards is missing or stale. Google Cloud's guidance is direct: if you want to connect your model with world knowledge, a wide range of topics, or up-to-date information on the internet, use Grounding with Google Search, which grounds responses in publicly available web data and returns Search Suggestions with the response so readers can follow the source. That matches every constraint here: public information, constantly changing, sources required (option D).
- A is wrong because few-shot examples shape format and tone. They teach style, not this week's visa rules. - B is wrong because a quarterly fine-tune is slow, expensive, and still leaves a stale window, and the border reopening from last week is not in the editorial team's back catalogue anyway. - C is wrong because temperature controls randomness in word choice, not factual currency. A low-temperature model states outdated visa rules just as confidently.
Reference: Grounding with Google Search