Training cutoff
The approximate date after which a model's main training data stops, stated in its documentation. Material created after the cutoff is a ready source of clean items the model is very unlikely to have memorized, making it a cheap contamination probe, though the date is approximate and a later fine-tuning stage can still introduce newer material.
Defined in 2 GAGE programs, which carry 3 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.
The approximate date after which a model's main training data stops, stated in its documentation. Material created after the cutoff is a ready source of clean items the model is very unlikely to have memorized, making it a cheap contamination probe, though the date is approximate and a later fine-tuning stage can still introduce newer material.
The point in time after which a model has no built-in knowledge of events, because its training data ends there. A primary source of outdated-fact errors.
The fixed date up to which a model's training data extends. The model has no knowledge of events, rulings, releases, or facts after this date and will often fabricate a plausible answer rather than admit the gap when asked about recent matters.
Where it is taught
The exact lessons this term appears in. The first module of every program is free with a free account.
- Evaluating AI Outputs for Accuracy · Practical AI Workflow Design and Prompt Engineering, Certified AI Practitioner: Workplace Foundations
- What a model cannot know: hallucination produced on demand, then caught · Build Before You Govern, Certified AI Governance Professional (CAIGP)
- Reproducing a claim: testing a vendor benchmark yourself in an afternoon · Staying Current: The Frontier Discipline, Certified AI Governance Professional (CAIGP)
Where it shows up at work
A definition is where Training cutoff 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.
Recognizes hallucination, drift, skew, brittleness and biased outcomes, and knows how each one enters a system.
2 lessons on this page teach it as a primary outcome.
Designs tests for factuality, robustness, fairness, safety and abuse resistance with rubrics, baselines and thresholds, and says what a score misses.
One lesson on this page teach it as a primary outcome.
Uses generative AI tools daily, knows what a model can and cannot do, and can say where an output should not be trusted.
One lesson 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.
- Model Risk Managerneeds it at depth
- AI Evaluation Specialistneeds it at depth · 6 open postings this week
- AI Model Validatorneeds it at depth
- Ethical AI Specialistneeds it at depth · 5 open postings this week
- AI Literacy Leadneeds it at depth
- Operations Manager, AI-readyneeds it at depth
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: NIST AI RMF MEASURE 2; ISO/IEC 42001; NIST AI RMF MEASURE 1 and 2; NIST AI 600-1.
One interview question about it
Tell me about a time a model was confidently wrong. How did you notice, and what did you change afterwards?
Asked of candidates for model failure modes and bias recognition. 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.
- A data science, statistics or analytics degree carries it as work you can show: Bias and failure modes were in your projects, not only your reading.
- Model risk, credit risk and quantitative analysis carries it as work you can show: Test design and benchmark selection are the craft.
- A computer science or software engineering degree carries it as a claim to back up: Testing is second nature; AI evaluation design is a new shape of 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.