Thecastle

https://atlas.mitre.org
MITRE ATLAS es una base de conocimiento sobre t cticas y t cnicas de adversarios para la seguridad en inteligencia artificial

https://www.sans.org › white-papers › automating...
Automated testing during the build stage of the AI engineering life cycle can evaluate the effectiveness of generative AI

https://www.sans.edu › cyber-research › automating...
Automated testing during the build stage of the AI engineering life cycle can evaluate the effectiveness of generative AI

https://www.articsledge.com › post › mitre-atlas-ai-threat-framework
As large language models moved into production MITRE progressively added generative AI tactics and techniques

https://www.cyberhaven.com › infosec-essentials › mitre-atlas
MITRE ATLAS classifies prompt injection as a technique in which an adversary embeds hidden instructions in text

https://learn.microsoft.com › ... › prompt-injection
Describes direct and indirect prompt injection attacks against LLMs why they succeed and the controls needed to

https://www.cohesity.com › platform › redlab › advisories › ai-prompt-injection
This advisory explains how these attacks work in practice maps them to MITRE ATLAS techniques AML T0051 AML T0093 and

https://github.com › PromptStrike › AI-security-prompt...
Structured prompt corpus for defensive AI security testing covering semantic drift prompt injection RAG exploitation embedding

https://cymulate.com › blog › mitre-atlas-ai-security-framework
An attack involving an AI system may combine ATLAS techniques such as prompt injection or data poisoning with