{
 "id": "https://redfinchadvisory.com/lexicon/ai-security",
 "slug": "ai-security",
 "prefLabel": "AI security",
 "altLabel": [
  "Security for AI",
  "LLM security"
 ],
 "hiddenLabel": [],
 "definition": "The practice of defending an AI system, and the data and tools it can reach, from attackers and deliberate misuse.",
 "scopeNote": "AI security is concerned with defending a system from misuse, especially intentional misuse that could jeopardize the model, its data, its tools or its outputs. AI safety covers harm with no attacker, including accidents.",
 "tier": "reference",
 "group": "Risks",
 "broader": [],
 "narrower": [
  "https://redfinchadvisory.com/lexicon/data-exfiltration",
  "https://redfinchadvisory.com/lexicon/indirect-prompt-injection",
  "https://redfinchadvisory.com/lexicon/jailbreak",
  "https://redfinchadvisory.com/lexicon/prompt-injection"
 ],
 "related": [
  {
   "id": "https://redfinchadvisory.com/lexicon/ai-safety",
   "type": "pair"
  }
 ],
 "links": [
  {
   "id": "https://redfinchadvisory.com/lexicon/red-teaming",
   "label": "tests",
   "direction": "in"
  },
  {
   "id": "https://redfinchadvisory.com/lexicon/guardrail",
   "label": "defended by",
   "direction": "out"
  }
 ],
 "modified": "2026-10-21",
 "markdown": "https://redfinchadvisory.com/lexicon/ai-security.md",
 "inScheme": {
  "id": "https://redfinchadvisory.com/lexicon",
  "version": "1.0",
  "currentAsOf": "2026-10-21"
 }
}