Mitigating Image-Based Prompt Injection in Multimodal AI

AI-103 By islever Editorial Last updated

Implement responsible AI for multimodal content

Short answer

B Configuring protected material detection does not mitigate risks from unsafe images or embedded hidden instructions as it only identifies copyrighted content in outputs.

Question

Note: This section contains one or more sets of questions with the same scenario and problem. Each question presents a unique solution to the problem. You must determine whether the solution meets the stated goals. More than one solution in the set might solve the problem. It is also possible that none of the solutions in the set solve the problem.
After you answer a question in this section, you will NOT be able to return. As a result, these questions do not appear on the Review Screen.
You have a multimodal AI generative model that accepts image uploads and uses extracted image text to generate responses.
You discover that users can upload unsafe images and embed hidden instructions into images to manipulate the model.
You need to implement controls to mitigate the risk.
Solution: You configure protected material detection.
Does this meet the goal?
  1. Yes
  2. No Correct Answer

Key Takeaways

The exam tests the distinction between copyright protection mechanisms and content safety filters, with the common trap being the assumption that all 'detection' features protect against malicious inputs.

This question addresses the use of protected material detection to mitigate risks from unsafe images and hidden instructions in multimodal models. It establishes that this specific control is incorrect for the stated security goal.

Common mistake: Candidates often select 'Yes', incorrectly assuming that protected material detection serves as a general-purpose safety filter for image uploads or prompt injection attacks.

Expert Analysis

Why the Answer Is Correct

Protected material detection is designed specifically to identify copyrighted or owned content within the model's output, ensuring compliance with intellectual property rights. It does not analyze input images for malicious payloads, hidden instructions (prompt injection), or unsafe content categories like violence or hate speech. Therefore, configuring this feature fails to mitigate the risk described in the scenario.

Why the Other Options Are Wrong

Since Option A ('Yes') is incorrect, Option B ('No') is the only valid choice. The solution proposed does not meet the goal because it targets the wrong threat vector (copyright infringement vs. adversarial input manipulation).

Community Comment Notes

Community consensus strongly supports Option B. As RICHARDALEX007 noted, "Protected material detection identifies copyrighted/owned content in output. It has nothing to do with unsafe images..." Munish_27 also clarified that it is "for copy right related prevention, not for required protection." Dinesh_87 emphasized that it checks if output matches known copyrighted text, which is irrelevant to input safety.

Exam Strategy

When asked about responsible AI controls, carefully distinguish between output filtering (e.g., toxicity, PII, copyright) and input validation (e.g., prompt injection, malware). Always map the control name to its specific vendor documentation purpose rather than guessing based on the word 'protection'.

Frequently Asked Questions

What controls should be used instead?

Use content safety filters, toxic content detection, and input validation pipelines to detect and block unsafe images or prompt injections before they reach the model.

What is protected material detection?

It is a feature that scans model outputs for copyrighted text or code to prevent intellectual property violations, not for input security.

More AI-103 FAQ →

Discussion 3 contributions

Compiled from public community threads and reader submissions, lightly edited for clarity. Handles and vote tallies are shown as recorded, and comments reflect the views of individual contributors.

From public community threads 3

  1. RICHARDALEX007 👍 2 Answered B
    Protected material detection identifies copyrighted/owned content in output. It has nothing to do with unsafe images or embedded malicious instructions, so it does not meet the goal.
  2. Dinesh_87 👍 1 Answered B
    Protected material detection serves a completely different purpose — it's designed to identify copyrighted or owned content, specifically pinpointing copyrighted content and providing sources for preexisting text and code. It checks whether the model's output matches known copyrighted text or code (song lyrics, articles, licensed code snippets, etc.) To actually meet the goal, you'd need: Content Safety image/multimodal harm filters → for unsafe images Prompt Shields for Documents → for hidden instructions in extracted image text
  3. Munish_20 👍 1 Answered B
    protected material detection is for copy right related prevention, not for required protection

Reader comments & corrections

No reader corrections yet — spotted an error or have a note? Share it below.

Log in to comment, report an error, or add a note about this question.

Which Request Parameters Maximize Foundry Output Stability and Reasoning?

S-grade

Azure Custom Vision: ordered steps to detect factory component faults

S-grade

Improving Agent Response Completeness in Microsoft Foundry

S-grade

Restricting Access to Microsoft Foundry Service via VNet

S-grade

Practice All AI-103 Questions

Access 88 questions with complete answers and detailed explanations.

View Full AI-103 Practice Test →

← Back to AI-103 Study Guide