# AI training data glossary

URL: https://aidataforsale.com/glossary
Publisher: AI Data For Sale (Marketing Signals Ltd)
Updated: 2026-11-10

Definitions of terms used in AI training data licensing.

- **[AI training data](https://aidataforsale.com/glossary/ai-training-data)**: The examples a machine learning model learns from, such as text, images, audio, video, code, actions and human judgments.
- **[Data license](https://aidataforsale.com/glossary/data-license)**: A contract granting a named party defined rights to use a copy of data, for set purposes and terms, while the licensor keeps ownership.
- **[Exclusive license](https://aidataforsale.com/glossary/exclusive-license)**: A license under which only the licensee may use the data for the agreed scope and term, so the owner cannot license it to others.
- **[Non-exclusive license](https://aidataforsale.com/glossary/non-exclusive-license)**: A license that lets the owner grant the same rights to other licensees.
- **[Sublicensing](https://aidataforsale.com/glossary/sublicensing)**: When a licensee grants rights in licensed data to a further party, such as a customer or contractor.
- **[Data provenance](https://aidataforsale.com/glossary/data-provenance)**: The documented history of a dataset: who created it, how it was collected, under what rights, and what was done to it since.
- **[Data manifest](https://aidataforsale.com/glossary/data-manifest)**: An inventory describing a dataset's contents, sources, volumes, date ranges and rights status, without containing the data itself.
- **[Dataset card](https://aidataforsale.com/glossary/dataset-card)**: A structured document describing a dataset's purpose, composition, collection, processing and intended uses.
- **[Chain of custody](https://aidataforsale.com/glossary/chain-of-custody)**: The record of every party that has held a copy of a dataset, and when, from creation to the current holder.
- **[De-identification](https://aidataforsale.com/glossary/de-identification)**: Removing or replacing information that identifies people so data cannot reasonably be linked back to them under a stated standard.
- **[Pseudonymization](https://aidataforsale.com/glossary/pseudonymization)**: Replacing identifiers with consistent codes so records can be linked to each other but not to a person without a separate key.
- **[Anonymization](https://aidataforsale.com/glossary/anonymization)**: Processing data so no one can identify the people in it by any reasonably likely means.
- **[Re-identification](https://aidataforsale.com/glossary/re-identification)**: Linking de-identified data back to the people it describes, often by combining it with other data.
- **[HIPAA safe harbor](https://aidataforsale.com/glossary/hipaa-safe-harbor)**: A HIPAA de-identification method that removes 18 specified types of identifiers from health information.
- **[Expert determination](https://aidataforsale.com/glossary/expert-determination)**: A HIPAA de-identification method where a qualified expert certifies that the risk of re-identifying individuals is very small.
- **[Protected health information (PHI)](https://aidataforsale.com/glossary/protected-health-information)**: Individually identifiable health information held by HIPAA covered entities and their business associates.
- **[Personally identifiable information (PII)](https://aidataforsale.com/glossary/pii)**: Information that can identify a person directly, such as a name or email, or indirectly when combined with other data.
- **[Pre-training](https://aidataforsale.com/glossary/pre-training)**: The first stage of training a foundation model on very large, broad datasets to learn general knowledge and language.
- **[Fine-tuning](https://aidataforsale.com/glossary/fine-tuning)**: Further training a pre-trained model on a smaller, targeted dataset to improve it at specific tasks or domains.
- **[Supervised fine-tuning (SFT)](https://aidataforsale.com/glossary/supervised-fine-tuning)**: Fine-tuning on examples of inputs paired with the desired outputs, often written or checked by people.
- **[RLHF (reinforcement learning from human feedback)](https://aidataforsale.com/glossary/rlhf)**: Training a model to produce responses people prefer, using human rankings or ratings of model outputs as the learning signal.
- **[Preference data](https://aidataforsale.com/glossary/preference-data)**: Records of which of several model responses a person judged better, often with reasons or rewrites.
- **[Evaluation dataset](https://aidataforsale.com/glossary/evaluation-dataset)**: A set of tasks with checked answers used to measure a model's ability, kept separate from training data.
- **[Hold-out set](https://aidataforsale.com/glossary/hold-out-set)**: Data deliberately kept out of training so it can test whether a model generalizes to examples it has not seen.
- **[Benchmark contamination](https://aidataforsale.com/glossary/benchmark-contamination)**: When test or benchmark data ends up in a model's training data, inflating its scores without improving real ability.
- **[RL environment](https://aidataforsale.com/glossary/rl-environment)**: A simulated or copied software setting where an AI agent can attempt tasks repeatedly and be scored, so it learns by trial and error.
- **[Agent trajectory](https://aidataforsale.com/glossary/agent-trajectory)**: A step-by-step record of actions taken to complete a task in software, including screens, inputs and results.
- **[Synthetic data](https://aidataforsale.com/glossary/synthetic-data)**: Data generated by models or simulations rather than collected from real people or systems.
- **[Model collapse](https://aidataforsale.com/glossary/model-collapse)**: Degradation in model quality when models are trained repeatedly on data generated by other models instead of human data.
- **[Success fee](https://aidataforsale.com/glossary/success-fee)**: A fee paid to an intermediary only when a transaction completes, usually a share of the deal value.
- **[Data broker](https://aidataforsale.com/glossary/data-broker)**: In privacy law, a business selling personal data about people it has no relationship with; in AI data, an intermediary matching data holders with buyers.
- **[Fair use](https://aidataforsale.com/glossary/fair-use)**: A US copyright doctrine allowing limited use of copyrighted material without permission, judged case by case.