MLchartDataset catalogue

jailbreak

Term · Cybersecurity · MLC-T-CYB-002281

A direct prompting attack intended to circumvent restrictions placed on model outputs, such as circumventing refusal behaviour to enable misuse.

Table 1. Record
IdentifierMLC-T-CYB-002281
FieldCybersecurity
ReferencesNIST AI 100-2e2025; NIST CSRC Glossary
Record as JSON 
{
  "id": "MLC-T-CYB-002281",
  "term": "jailbreak",
  "field": "Cybersecurity",
  "definition": "A direct prompting attack intended to circumvent restrictions placed on model outputs, such as circumventing refusal behaviour to enable misuse.",
  "references": [
    "NIST AI 100-2e2025",
    "NIST CSRC Glossary"
  ],
  "url": "https://mlchart.com/terminology/cybersecurity/jailbreak/"
}

Record 2,281 of 4,693 in Cybersecurity terminology (MLC-0102). Request the full dataset.