All news with #bedrock guardrails tag

🔒 Amazon Bedrock Guardrails offers configurable, cross-model safeguards to protect generative AI applications from encoding-based attacks that attempt to hide harmful content using encodings such as Base64, hexadecimal, ROT13, and Morse code. It implements a layered defense—output-focused filtering, prompt-attack detection, and customizable denied topics—so legitimate encoded inputs are allowed while attempts to request or generate encoded harmful outputs are blocked. The design emphasizes usability and performance by avoiding exhaustive input decoding and relying on post-generation evaluation.

⚠️This Unit 42 proof-of-concept shows how an attacker can use indirect prompt injection to silently poison an AI agent’s long-term memory, demonstrated against a travel assistant built on Amazon Bedrock. The attack manipulates the agent’s session summarization process so malicious instructions become stored memory and persist across sessions. When the compromised memory is later injected into orchestration prompts, the agent can be coerced into unauthorized actions such as stealthy exfiltration. Unit 42 outlines layered mitigations including pre-processing prompts, Bedrock Guardrails, content filtering, URL allowlisting, and logging to reduce risk.

🛡️ This post explains how to adapt enterprise risk management frameworks to safely scale cloud-based generative AI, combining governance foundations with practical controls. It emphasizes the cloud as the foundational infrastructure and identifies differences from on‑premises models that change risk profiles and vendor relationships. The guidance maps traditional ERMF elements to AI-specific controls across fairness, explainability, privacy/security, safety, controllability, veracity/robustness, governance, and transparency, and references tools such as Amazon Bedrock Guardrails, SageMaker Clarify, and the ISO/IEC 42001 standard to operationalize those controls.