{
 "id": "https://redfinchadvisory.com/lexicon/reward-hacking",
 "slug": "reward-hacking",
 "prefLabel": "Reward hacking",
 "altLabel": [],
 "hiddenLabel": [],
 "definition": "A failure where a model or agent games how it is measured, getting a high score without the intended outcome, such as editing tests so they pass instead of fixing the code.",
 "scopeNote": "",
 "tier": "reference",
 "group": "Risks",
 "broader": [],
 "narrower": [],
 "related": [
  {
   "id": "https://redfinchadvisory.com/lexicon/eval",
   "type": "related"
  },
  {
   "id": "https://redfinchadvisory.com/lexicon/regression-eval",
   "type": "related"
  },
  {
   "id": "https://redfinchadvisory.com/lexicon/alignment",
   "type": "related"
  }
 ],
 "links": [],
 "modified": "2026-10-21",
 "markdown": "https://redfinchadvisory.com/lexicon/reward-hacking.md",
 "inScheme": {
  "id": "https://redfinchadvisory.com/lexicon",
  "version": "1.0",
  "currentAsOf": "2026-10-21"
 }
}