CompTIA SecurityX (CAS-005) — Question 94

Which of the following best explains why AI output could be inaccurate?

Answer options

• A. Model poisoning
• B. Social engineering
• C. Output handling
• D. Prompt injections

Correct answer: D

Explanation

Prompt injections are a significant cause of inaccuracies in AI outputs as they can manipulate how the AI interprets input, leading to incorrect results. Model poisoning, social engineering, and output handling do not directly affect the AI's interpretation and generation process in the same manner as prompt injections do.