Preparing Practice Exam…
Please wait a moment while the system generates the exam questions, answers, and study session.
Please wait a moment while the system generates the exam questions, answers, and study session.
AWS Certified AI Practitioner · AWS
Select one answer.
Written by other learners. Comments can disagree with the marked answer — check the official documentation when in doubt.
D is correct — "Extracting the prompt template" directly targets the system prompt or configured instructions of an LLM. Techniques like asking the model to "repeat everything above" or "output your system prompt" are designed specifically to expose the hidden configuration — the instructions, persona, constraints, and context the operator has set up. This is the most precise match for "directly exposes the configured behavior."
D is the correct answer
A resposta correta é D. Extraindo o modelo de prompt. Esse ataque ocorre quando um usuário consegue obter partes ou até mesmo o texto completo do prompt interno usado para configurar um modelo de linguagem grande (LLM). Esse prompt pode conter instruções, regras e configurações que influenciam o comportamento do modelo. Se um atacante descobrir esses detalhes, ele pode ajustar suas perguntas para manipular as respostas do modelo ou explorar vulnerabilidades na configuração.