Syllaro Academy

AWS Certified AI Practitioner

Prompt Attacks and Risks

How attackers turn your own prompt against you: prompt injection, jailbreaking, and prompt leaking, plus the layered defenses including Amazon Bedrock Guardrails.

Intermediate 18 minutes 4 Learning Objectives
  1. Explain prompt injection and why it resembles SQL injection
  2. Distinguish prompt injection, jailbreaking, and prompt leaking
  3. Describe layered defenses: input isolation, instruction hardening, and Amazon Bedrock Guardrails
  4. Identify who owns prompt-security risk under the AWS shared responsibility model