Google Cloud Generative AI Leader GAIL — Question 60

The correct answer is A. Option A: To filter out potentially harmful or inappropriate content from the model’s output based on the desired level of filtering. The other options (B, C, D) are less suitable because they add operational overhead, cost, or complexity, or they do not fully satisfy the stated requirement.