Hallucination
The phenomenon in which an AI system, particularly a large language model, produces fluent, confident, but factually incorrect or entirely fabricated content, because the system generates statistically likely text rather than verifying truth against a fact database. Hallucination is close to an inherent property of how these systems currently generate output, not a rare malfunction, though techniques like retrieval-augmented generation can reduce its frequency.
Defined in 5 GAGE programs, which carry 12 distinct definitions of it. The wording above is taught in The AI Lobbyist: Certified AI Policy Strategist.
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.
The phenomenon in which an AI system, particularly a large language model, produces fluent, confident, but factually incorrect or entirely fabricated content, because the system generates statistically likely text rather than verifying truth against a fact database. Hallucination is close to an inherent property of how these systems currently generate output, not a rare malfunction, though techniques like retrieval-augmented generation can reduce its frequency.
In generative AI, an output that is fluent and confident but not supported by the source or by fact, for example a summary that states a medication the source note never mentions. The term is imprecise until defined for a specific task; a defensible metric replaces the word with an exact, gradable rule.
When an AI system states something false with full confidence, because it generates the most plausible output rather than the true one and cannot tell the difference. Not a bug awaiting a patch but a property of the technology, which is why high-stakes uses need a verification layer.
When an AI states something false with confidence, such as inventing a statistic, source, or date it does not actually have. Instructing the AI to admit uncertainty is a first-line guardrail against acting on hallucinated information.
Confident, fluent, well formatted false output. Expected behaviour of a system fitted to produce plausible material, not a malfunction awaiting a fix.
Where it is taught
The exact lessons this term appears in. The first module of every program is free with a free account.
- Building Your AI Briefcase · Your AI Learning Companion, Certified AI Practitioner: Workplace Foundations
- What is AI? Definitions, History, and Types · AI Fundamentals, Certified AI Practitioner: Workplace Foundations
- How AI Actually Works: Algorithms, Data, and Training · AI Fundamentals, Certified AI Practitioner: Workplace Foundations
- Building the eval suite that would have caught your Module 3 incident · Evaluation and Trust, Certified AI Governance Professional (CAIGP)
- The Only Ten Things About AI That Matter in Policy · The Point Person Mandate, The AI Lobbyist: Certified AI Policy Strategist
- The Five Analogies That Always Work with Lawmakers · The Point Person Mandate, The AI Lobbyist: Certified AI Policy Strategist
- What AI Really Is: No Hype, No Myths, No Fear · Foundations of AI for Lobbyists, The AI Lobbyist: Certified AI Policy Strategist
- Opposition Research Automation on the Public Record · Prompt Engineering for Strategic Persuasion, The AI Lobbyist: Certified AI Policy Strategist
- The AI Technology Stack Decoded · AI Literacy for Decision Makers, Certified AI Transformation Professional (CATP)
- AI Risk Management · Governance and Responsible AI, Certified AI Transformation Professional (CATP)
Where it shows up at work
A definition is where Hallucination 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.
Uses generative AI tools daily, knows what a model can and cannot do, and can say where an output should not be trusted.
2 lessons on this page teach it as a primary outcome.
Explains training, tokens, context windows, embeddings, retrieval and fine-tuning well enough to ask an engineer a precise question and spot weak evidence.
2 lessons on this page teach it as a primary outcome.
Builds production systems on foundation models: retrieval augmented generation, structured tool use, evaluation harnesses, guardrails, and cost and latency budgets.
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 Adoption and Enablement Leadneeds it at depth
- Operations Manager, AI-readyneeds it at depth
- AI Literacy Leadneeds it at depth
- AI Engineer, production systems editionneeds it at depth · 23 open postings this week
- Chief Technology Officer, AI engineering editionneeds it at depth
- AI Security Architectneeds it at depth · 8 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 Article 4; DOL AI Literacy Framework; NIST AI RMF.
One interview question about it
Walk me through a task you now do with an AI tool. Where did you stop trusting its output, and how did you know?
Asked of candidates for working ai fluency. 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.
If you come from another field
Some backgrounds already carry this competency, and the readiness check gives credit for it.
- Recent graduate in law, policy, business or data carries it as a claim to back up: Most graduates already use the tools daily.
- Teachers, trainers and instructional designers carries it as a claim to back up: Most educators are already using AI in preparation.
- Data analysis, stewardship and business intelligence carries it as a claim to back up: SQL and analytical fluency make the technical layer approachable.
Terms it appears with
Not an alphabetical neighbourhood: these are the terms taught in the same lessons, ranked by how often they appear together.