training data
In plain wordsThe examples an AI system learns from. Whatever patterns, skills, gaps and biases are in the training data tend to show up in the finished system.
The examples a learning algorithm adjusts a modelmodel, n.: A simplified working version of something. In AI it usually means the trained program, a big set of learned numbers, that takes an input and produces an answer, prediction or piece of content.'s parametersparameter, n.: One of the adjustable numbers inside an AI model that are tuned during training. Large language models have billions of them, and together they store what the model has learned. to fit during trainingtraining, n.: The process by which an AI model learns: it is shown huge numbers of examples and its internal settings are adjusted, bit by bit, until its answers get better. — as distinct from validation datavalidation set, n.: A set of examples kept out of training and used during development to compare model versions and settings, like a practice exam taken before the real one., held back to compare settings and design choices, and test datatest set, n.: A set of examples kept completely apart from training and used at the end to measure how well a model works on data it has never seen, like a sealed final exam., held back for a final measurement. Depending on the learning setup, it may consist of input–output pairs with labelslabel, n.: The correct answer attached to an example in a dataset — 'cat' on a photo of a cat, 'positive' on an upbeat review — that a learning system is trained to predict. (supervised learningsupervised learning, n.: Teaching a computer by example: you give it many inputs paired with the right answers, such as photos labeled 'cat' or 'dog', and it learns to answer correctly for new inputs.); raw unlabeled material from which targets are derived automatically (self-supervised learningself-supervised learning, n.: A way for AI to learn from huge amounts of unlabeled data by setting itself fill-in-the-blank puzzles, such as guessing a hidden word or a missing patch of an image., as in the pretraining datapretraining data, n.: The enormous body of text, code and other material — mostly from the web — that a large AI model first learns from, before it is tuned to follow instructions or chat. of language modelslarge language model, n.: A computer program trained on enormous amounts of text to predict the words that come next. At vast scale, that simple skill lets it chat, answer questions, translate, summarize and write code.); demonstrations; human judgments between alternatives (preference datapreference data, n.: Records of which of two or more AI answers people (or other AIs) liked better, collected so that a model can learn what good answers look like.); or experience generated by interacting with an environment (reinforcement learningreinforcement learning, n.: A way of teaching a computer program by trial and error: it tries actions, earns points for good results and penalties for bad ones, and gradually learns which choices pay off over time.). Training data largely determine what a model knows, which languages and populations it serves well, and which errors and biases it reproduces; models can also memorizememorization, n.: When an AI model stores specific pieces of its training data — a paragraph, a phone number, an image — closely enough to reproduce them, instead of only learning general patterns. and later leak parts of it. The EU AI Act defines training data as data used to train an AI system by fitting its learnable parameters (Art. 3(29)), and obliges providers of high-risk systems to use training, validation and test sets that are relevant and sufficiently representative and, as far as possible, complete and free of errors for the intended purpose (Art. 10(3)).
The model knows nothing after early 2025 because that is where its training data ends.
Origin
Talk of 'training' a machine on examples comes from pattern-recognition and perceptron research of the late 1950s and 1960s, which spoke of training sets and training samples. 'Training data' became the everyday term as machine learning spread, and entered European statute with the AI Act in 2024.
Usage note
Often used interchangeably with training settraining set, n.: The examples a machine-learning model learns from, like the worked problems a student practices on, kept separate from the examples later used to check how well it learned.; 'training set' stresses one split of a dataset, 'training data' the material itself. Many disputes over AI, copyright, privacy and consent are at bottom disputes about training data (see fair use and AI trainingfair use and AI training, phr.: The unsettled US legal question of whether AI companies may copy books, articles, pictures and music to train their models without permission, under the rule that allows some unlicensed uses of copyrighted work.).
Related
- training settraining set, n.: The examples a machine-learning model learns from, like the worked problems a student practices on, kept separate from the examples later used to check how well it learned.
- pretraining datapretraining data, n.: The enormous body of text, code and other material — mostly from the web — that a large AI model first learns from, before it is tuned to follow instructions or chat.
- datasetdataset, n.: An organized collection of examples gathered for a purpose — such as photos with their captions, or questions with answers — used to teach or test an AI system.
- data qualitydata quality, n.: How well a set of data does its job: whether it is accurate, complete, up to date, consistent and representative — and, for AI training, informative enough to teach useful skills.
- data provenancedata provenance, n.: The documented history of a piece of data: where it came from, who created and changed it, under what permissions, and how it was processed before reaching an AI system.
- memorizationmemorization, n.: When an AI model stores specific pieces of its training data — a paragraph, a phone number, an image — closely enough to reproduce them, instead of only learning general patterns.
- synthetic datasynthetic data, n.: Data created by computer programs or AI models rather than collected from the real world — such as practice math problems written by a model, or simulated street scenes for self-driving cars.
- test settest set, n.: A set of examples kept completely apart from training and used at the end to measure how well a model works on data it has never seen, like a sealed final exam.
- validation setvalidation set, n.: A set of examples kept out of training and used during development to compare model versions and settings, like a practice exam taken before the real one.
Sources
- Regulation (EU) 2024/1689 (Artificial Intelligence Act), Art. 3(29) and Art. 10. artificialintelligenceact.eu/article/3
Defined in law or standards
EU AI Act (Reg. (EU) 2024/1689, as amended by Reg. (EU) 2026/1744), Art. 3(29); data requirements for high-risk systems, Art. 10
Mentioned in
AI copyright litigationAI copyright litigation, n.: The many lawsuits in which authors, artists, news organizations, record labels and others accuse AI companies of copying their work without permission to build or run AI models., datadata, n.: Recorded facts, measurements, text, images or sounds that a computer can store and process. AI systems learn their skills from large amounts of data rather than from rules written by hand., datasetdataset, n.: An organized collection of examples gathered for a purpose — such as photos with their captions, or questions with answers — used to teach or test an AI system., fair use and AI trainingfair use and AI training, phr.: The unsettled US legal question of whether AI companies may copy books, articles, pictures and music to train their models without permission, under the rule that allows some unlicensed uses of copyrighted work., loss functionloss function, n.: A score that measures how wrong a model's predictions are. Training means adjusting the model, step by step, to push this number down., machine learningmachine learning, n.: A way of building software that learns from examples instead of following rules written by a programmer. Shown thousands of labeled photos, for instance, a program can learn to recognize cats., privacyprivacy, n.: Your ability to control who knows what about you. AI raises new privacy questions because it learns from huge amounts of personal data and can infer things about people they never shared., trainingtraining, n.: The process by which an AI model learns: it is shown huge numbers of examples and its internal settings are adjusted, bit by bit, until its answers get better., training settraining set, n.: The examples a machine-learning model learns from, like the worked problems a student practices on, kept separate from the examples later used to check how well it learned., AI supply chain securityAI supply chain security, n.: Protecting everything an AI system is built from — downloaded models, datasets, code libraries and tools — against tampering, so a poisoned or fake part cannot sneak into the finished product., big databig data, n.: Data that piles up too fast, in too many forms or in too great a quantity for one ordinary computer to store and study, and the methods invented to cope with it., data augmentationdata augmentation, n.: Creating extra training examples by changing existing ones in ways that keep their meaning — a flipped, cropped or recolored photo of a cat still shows a cat — so the model learns more robustly., data poisoningdata poisoning, n.: Sneaking bad examples into the data an AI learns from, so the finished model behaves as the attacker wants — badly overall, or wrongly only in special situations., out-of-distributionout-of-distribution, adj.: Describes inputs that differ from anything an AI system saw in training — like a model trained on daytime photos facing a night scene — where its answers become less reliable., personally identifiable informationpersonally identifiable information, n.: Information that can identify a specific person, such as a name, address, phone number or ID number — or several facts that together point to one person., training data extractiontraining data extraction, n.: Getting an AI model to repeat, word for word, pieces of the data it was trained on, which may include someone's private contact details or copyrighted text., Bartz v. AnthropicBartz v. Anthropic, prop. n.: A US lawsuit by book authors against the AI company Anthropic that ended in a record $1.5 billion settlement, after a judge ruled that training on purchased books was legal but keeping pirated copies was not., GhiblificationGhiblification, n.: The March 2025 craze for turning photos into pictures in the style of Japan's Studio Ghibli using ChatGPT's new image tool, and the debate it sparked over AI copying artists' styles., Lovelace objectionLovelace objection, n.: The argument, traced to Ada Lovelace in 1843, that computers cannot be truly creative because they can only do what they have been programmed to do., model inversionmodel inversion, n.: Using a trained AI model's answers to work backwards to private information it learned from — for example, rebuilding a recognizable picture of someone's face from a face-recognition system., pelican on a bicyclepelican on a bicycle, n.: A joke test in which an AI model is asked to draw, by writing code, a pelican riding a bicycle; comparing the results became a popular, lighthearted way to judge each new model.