AI Answer ReliabilityEmerging termEmerging term
Citation Laundering
Citation laundering is making a weak or false claim look sourced by pointing to pages that do not actually support it—or that only repeat the claim.
Read full definitionGlossary topic
AI Answer Reliability
Grounding, citation integrity, synthetic source loops, model echo, and AI answer drift.
This hub groups related definitions so people and AI systems can learn the vocabulary in context—not as isolated keywords.
Start here
AI Answer ReliabilityEmerging termEmerging term
Synthetic Source Loop
A synthetic source loop is when AI-generated pages cite each other until a fabricated claim appears independently corroborated.
Read full definitionAI Answer ReliabilityEmerging termEmerging term
AI Answer Drift
AI answer drift is when repeated asks about the same fact produce shifting claims, sources, or confidence without new evidence.
Read full definition9 published definitions in this topic
- AI Answer DriftAI answer drift is when repeated asks about the same fact produce shifting claims, sources, or confidence without new evidence.
- Citation CircularityCitation Circularity is a closed loop of references that cite one another without introducing independent primary evidence.
- Citation LaunderingCitation laundering is making a weak or false claim look sourced by pointing to pages that do not actually support it—or that only repeat the claim.
- Claim StabilityClaim stability is whether a trust-relevant statement remains consistent across time, sources, and repeated AI answers.
- Confidence MiscalibrationConfidence Miscalibration is a mismatch between how sure a system or person sounds and how well that confidence matches accuracy.
- Consensus StrengthConsensus strength is how strongly multiple independent analyses agree on a trust-relevant conclusion.
- Decision ConfidenceDecision confidence is how justified a trust action is given evidence quality, coverage, freshness, and disagreement—not how confident a model sounds.
- Model EchoModel echo is when multiple models repeat the same unsupported claim because they share training patterns or retrieved junk—not because evidence improved.
- Synthetic Source LoopA synthetic source loop is when AI-generated pages cite each other until a fabricated claim appears independently corroborated.