AI-900 Practice Question: Describe features of generative AI workloads on Azure
What is 'prompt injection' and why is it a security concern for AI applications?
⚠ Common exam trap
Test-takers frequently confuse prompt injection with benign prompt engineering or testing activities, failing to recognize that the key distinction is malicious intent to override system instructions rather than legitimate modification or evaluation of prompts.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Malicious input that overrides an AI system's instructions to hijack its behaviour
Prompt injection is a security vulnerability where an attacker crafts input that overrides or bypasses the system-level instructions (system prompt) of an AI model, causing it to behave in unintended ways. This is a critical concern because generative AI models, especially large language models (LLMs), are designed to follow instructions in the prompt, and a malicious user can inject commands that hijack the model's behavior, potentially exposing sensitive data, generating harmful content, or performing unauthorized actions. In Azure AI services, this risk is mitigated through content filtering, input validation, and the use of metaprompt protections.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
When developers inject test prompts to evaluate model performance
Why it's wrong here
Running benign test prompts to evaluate a model is a standard QA and red-teaming practice, not a prompt-injection attack as the term is used in security. Prompt injection specifically refers to an adversarial, untrusted input that is crafted to override the model's system instructions and alter its behavior in a harmful way. The intention and the threat model are different: an evaluator controls the test to observe performance, whereas an attacker controls the prompt to break the application's intended policy.
- ✓
Malicious input that overrides an AI system's instructions to hijack its behaviour
Why this is correct
Prompt injection is the deliberate insertion of malicious instructions into user-supplied or retrieved content that the model processes as data. Because an LLM treats the system prompt and incoming text as the same token stream, a crafted phrase can override the original instructions and force the model to ignore guardrails, reveal sensitive information, or take unintended actions. This is the canonical exploitation of instruction-following behavior and is one of the most serious security risks for LLM-based applications.
- ✗
The process of adding new prompts to expand a model's capability
Why it's wrong here
Adding new prompts to expand what a model can do is prompt engineering or fine-tuning, not an attack. In prompt injection, the attacker does not enhance the model — instead, they exploit the model's existing instruction-following capability to divert it from its intended behavior. The malicious 'new prompt' is not a feature; it is an unauthorized command that the model mistakenly treats with the same authority as its original system instructions.
- ✗
Accidentally sending the wrong prompt to the model due to a software bug
Why it's wrong here
Accidentally sending the wrong prompt because of a software bug is an implementation defect, not a security exploit. Prompt injection is a deliberate adversarial action where the attacker crafts input specifically to override the system's instructions, not a coding mistake in how prompts are routed or dispatched. The distinction matters because injection attacks can occur even when the software is perfectly functional — the vulnerability lies in the model's tendency to obey instructions embedded in untrusted text.
Go deeper
Related to this question
Learn chapter
Azure Machine Learning Studio
Key term
Generative AI
Generative AI is a type of artificial intelligence that creates new content—like text, images, or code—by learning patterns from existing data.
Key term
Azure AI Services
Azure AI Services is a collection of pre-built, cloud-based artificial intelligence APIs and services that allow developers and IT professionals to integrate capabilities like vision, speech, language, and decision-making into applications without needing deep machine learning expertise.
About these practice questions
Courseiva writes every AI-900 question from scratch — 985 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or dumps. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AI-900 practice question is part of Courseiva's free Microsoft certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AI-900 exam.