Confabulation
An alternative term for hallucination, used in the United States National Institute of Standards and Technology Generative AI Profile (NIST AI 600-1, 2024) to name confidently stated but erroneous or fabricated content. Preferred by some researchers because "hallucination" borrows a clinical term for perceiving things that are not present, and NIST's own framing cautions that terms like "hallucination" risk implying the system has perceptions or intentions it does not have. Either word names the same behavior; neither implies the model is aware of what it is doing.
Defined in 3 GAGE programs, which carry 7 distinct definitions of it. The wording above is taught in Certified AI Governance Professional (CAIGP).
How each discipline defines it
The same term does different work depending on who is using it. These are the definitions as each program teaches them, unedited.
An alternative term for hallucination, used in the United States National Institute of Standards and Technology Generative AI Profile (NIST AI 600-1, 2024) to name confidently stated but erroneous or fabricated content. Preferred by some researchers because "hallucination" borrows a clinical term for perceiving things that are not present, and NIST's own framing cautions that terms like "hallucination" risk implying the system has perceptions or intentions it does not have. Either word names the same behavior; neither implies the model is aware of what it is doing.
the technically precise term the NIST Generative AI Profile (AI 600-1) uses for what is colloquially called AI "hallucination," a generative AI system producing fluent but factually incorrect or fabricated output.
The production of confidently stated but erroneous or false content by which users may be misled or deceived. A published term preferable to the colloquial one because it names where the control must sit.
The Generative AI Profile's term for a generative model producing confident, fluent output that is false, commonly called hallucination. Named as one of the twelve generative-AI risk categories because a confident false output can be more dangerous than an obvious error.
The established tendency of a generative AI model, including one that retrieves documents or calls tools on your behalf, to produce confident, plausible, false output; the reason every AI-surfaced red-team finding is a lead to verify, not a verdict (NIST AI 600-1, Generative AI Profile, 2024).
Where it is taught
The exact lessons this term appears in. The first module of every program is free with a free account.
- What a model cannot know: hallucination produced on demand, then caught · Build Before You Govern, Certified AI Governance Professional (CAIGP)
- NIST AI RMF as an operating system: mapping your organization onto govern, map, measure, manage · The World's Rulebooks, Certified AI Governance Professional (CAIGP)
- Your conformity file under attack: the red team finds what you missed · Adversarial Governance, Certified AI Governance Professional (CAIGP)
- The Big Four Frameworks: EU AI Act, NIST AI RMF, ISO/IEC 42001, OECD Principles · Foundations of AI for Lobbyists, The AI Lobbyist: Certified AI Policy Strategist
- NIST Translation Mastery: Control, One-Liner, Memo Paragraph · AI Law in Practice: The Compliance-to-Persuasion Toolkit, The AI Lobbyist: Certified AI Policy Strategist
- Artificial intelligence assisted governance, risk and compliance work · Governance, Risk and Compliance Technology, Automation, and Evidence Engineering, Certified Enterprise Governance, Risk and Compliance Professional (CEGP)
- Governing the artificial intelligence your own function uses · Governance, Risk and Compliance Technology, Automation, and Evidence Engineering, Certified Enterprise Governance, Risk and Compliance Professional (CEGP)
Where it shows up at work
A definition is where Confabulation starts. The lessons above teach it inside the competencies below, which are what job descriptions ask for and what the GAGE credential records. Every link here is drawn from the same careers map that builds a role's readiness check.
Compares the EU AI Act, NIST AI RMF, ISO/IEC 42001 and sector rules by intent and control objective, and says where they do not overlap.
2 lessons on this page teach it as a primary outcome.
Decides what a person affected by an AI decision must be told, how an output can be explained, and how they can challenge it.
2 lessons on this page teach it as a primary outcome.
Runs GOVERN, MAP, MEASURE and MANAGE as a cycle with evidence, builds current and target profiles, and applies the generative AI profile.
2 lessons on this page teach it as a primary outcome.
Roles that hire for it
Each role page shows the full competency set, what the seat decides, and a readiness check against your own evidence.
- AI Compliance Managerneeds it at depth · 2 open postings this week
- Senior AI Compliance Analystneeds it at depth
- Security Compliance Managerneeds it at depth · 4 open postings this week
- AI Compliance Specialistneeds working knowledge · 1 open posting this week
- Responsible AI Leadneeds it at depth · 7 open postings this week
- Ethical AI Specialistneeds working knowledge · 5 open postings this week
How job descriptions say it
Postings rarely use the textbook word. These are the phrases the careers map reads as asking for the competencies above.
Where the frameworks place it: EU AI Act; NIST AI RMF; ISO/IEC 42001; EU AI Act Articles 13 and 50.
One interview question about it
Map one control to the EU AI Act, NIST AI RMF and ISO/IEC 42001 at once, and tell me where the mapping breaks.
Asked of candidates for framework crosswalking without false equivalence. Answering it in your own words is exactly how a GAGE topic is passed, and the credential records the pass where an employer can check it.
Terms it appears with
Not an alphabetical neighbourhood: these are the terms taught in the same lessons, ranked by how often they appear together.