Q51 — AWS AIF-C01 Ch.1

Question 51 of 100 | ← Chapter 1

A company is leveraging generative AI capabilities to assess its solution's security responsibilities across various implementation scopes. Based on these scopes, which solution scope assigns the greatest security responsibility to the company?

• A. Using a third-party enterprise application with embedded generative AI functionality.
• B. Building an application using an existing third-party generative AI foundation model (FM).
• C. Fine-tuning an existing third-party generative AI foundation model (FM) using business-specific data.
• D. Building and training a generative AI model from scratch using customer-owned, business-specific data. ✓

Correct Answer: D. Building and training a generative AI model from scratch using customer-owned, business-specific data.

Explanation

This question tests understanding of security responsibilities across different generative AI solution scopes. In AI, building and training a model from scratch using proprietary business data places full responsibility on the company for the entire lifecycle and outputs. Option A involves minimal responsibility as it uses a third-party application; Option B entails limited responsibility; Option C increases responsibility relative to A and B due to fine-tuning. Option D—building and training from scratch—imposes the highest security responsibility on the company.