# AI Basics vocabulary

150 AI Basics terms, from Novice to Master. Start with the basics or jump to a level.

Updated: November 19, 2026

## Novice

- [AI Assistant](https://www.ask-codey.com/glossary/ai-assistant): An AI model packaged to help with tasks.
- [Artificial Intelligence](https://www.ask-codey.com/glossary/artificial-intelligence): Software that does tasks we think need human intelligence.
- [Chatbot](https://www.ask-codey.com/glossary/chatbot): A program you talk to in messages.
- [Generative AI](https://www.ask-codey.com/glossary/generative-ai): AI that creates new text, images, code or audio.
- [Large Language Model](https://www.ask-codey.com/glossary/large-language-model): An AI model trained on huge text to understand and write language.
- [Machine Learning](https://www.ask-codey.com/glossary/machine-learning): Programs that learn patterns from examples.
- [Model](https://www.ask-codey.com/glossary/model-ai): The trained system that turns input into output.
- [Prompt](https://www.ask-codey.com/glossary/prompt): The instruction or question given to an AI model.

## Apprentice

- [Accuracy](https://www.ask-codey.com/glossary/accuracy): The share of a classifier's guesses that are right: right guesses divided by all guesses.
- [AI API](https://www.ask-codey.com/glossary/ai-api): A web API for sending prompts to a model from code.
- [AI Coding Assistant](https://www.ask-codey.com/glossary/ai-coding-assistant): AI that suggests and writes code in your tools.
- [Baseline](https://www.ask-codey.com/glossary/baseline-ai): The simplest honest guess, such as always guessing the average, that a model has to beat to be worth using.
- [Binary Classification](https://www.ask-codey.com/glossary/binary-classification): Classification with exactly two possible answers, such as spam or not spam, or fraud or not fraud.
- [Categorical Feature](https://www.ask-codey.com/glossary/categorical-feature): A feature whose values are categories, not amounts, such as species or neighborhood; most models need it encoded as numbers first.
- [Classification](https://www.ask-codey.com/glossary/classification): Machine learning that guesses a category, one of a fixed set of answers such as spam or not spam, from examples that come with their answers.
- [Classifier](https://www.ask-codey.com/glossary/classifier): A model that does classification: it guesses a category for each example, such as spam or not spam.
- [Cluster](https://www.ask-codey.com/glossary/cluster-ai): A group of examples that clustering found to be similar to each other, such as customers who visit early and buy little.
- [Context Window](https://www.ask-codey.com/glossary/context-window): How much text a model can consider at once.
- [Decision Tree](https://www.ask-codey.com/glossary/decision-tree): A model that guesses by asking a chain of yes-or-no questions about the features, like a flowchart, ending at a leaf with the answer.
- [False Negative](https://www.ask-codey.com/glossary/false-negative): A miss: the model said no, but the answer is yes, such as a fraud the check let through.
- [False Positive](https://www.ask-codey.com/glossary/false-positive): A false alarm: the model said yes, but the answer is no, such as a normal payment flagged as fraud.
- [Feature](https://www.ask-codey.com/glossary/feature): A column a model looks at to make its guess, such as the temperature when guessing a day's sales.
- [Hallucination](https://www.ask-codey.com/glossary/hallucination): An AI confidently stating something false.
- [Heuristic](https://www.ask-codey.com/glossary/heuristic): A hand-written rule of thumb, such as "Saturdays are busy, so bake more", used instead of a model learned from data.
- [Inference](https://www.ask-codey.com/glossary/inference): Using a trained model to produce output.
- [Intercept](https://www.ask-codey.com/glossary/intercept): Where a line starts: its guess when every feature is 0, a starting amount the slopes build on.
- [Label](https://www.ask-codey.com/glossary/label-ai): The column a model learns to guess, such as the cups sold; scikit-learn calls it the target.
- [Majority Vote](https://www.ask-codey.com/glossary/majority-vote): Picking the answer most of a group chose, such as the label most of a new example's nearest neighbors share.
- [Multimodal Model](https://www.ask-codey.com/glossary/multimodal-model): A model that handles text plus images, audio or documents.
- [Probability](https://www.ask-codey.com/glossary/probability): How likely something is, from 0 (it never happens) to 1 (it always does); 0.8 means about 8 times in 10.
- [Prompt Engineering](https://www.ask-codey.com/glossary/prompt-engineering): Writing prompts that get reliable results.
- [Pythagorean Theorem](https://www.ask-codey.com/glossary/pythagorean-theorem): The rule that a right triangle's long side, squared, equals the other two sides squared and added: 3² + 4² = 5².
- [Regression](https://www.ask-codey.com/glossary/regression): Machine learning that guesses a number, such as tomorrow's loaves or a home's price, from examples that come with their answers.
- [Slope](https://www.ask-codey.com/glossary/slope): How much a line's guess goes up for each one-unit step in a feature, in the label's units: dollars per square foot, say.
- [System Prompt](https://www.ask-codey.com/glossary/system-prompt): The standing instruction that sets an AI's role and rules.
- [Temperature](https://www.ask-codey.com/glossary/temperature): A setting for how random or creative output is.
- [Test Data](https://www.ask-codey.com/glossary/test-data): The examples held back from a model while it learns, so its guesses can be scored on examples it never saw.
- [Token](https://www.ask-codey.com/glossary/token): A chunk of text, about three-quarters of a word.
- [Training Data](https://www.ask-codey.com/glossary/training-data): The examples a model learned from.
- [True Negative](https://www.ask-codey.com/glossary/true-negative): A no the model got right: it let through an example that really isn't the positive class, such as a normal payment.
- [True Positive](https://www.ask-codey.com/glossary/true-positive): A yes the model got right: it flagged an example that really is the positive class, such as a fraud it caught.

## Journeyman

- [Algorithmic Bias](https://www.ask-codey.com/glossary/algorithmic-bias): When a model's decisions or mistakes fall unfairly on some groups of people, often learned from past decisions in its training data.
- [Balanced Accuracy](https://www.ask-codey.com/glossary/balanced-accuracy): The average of each class's recall, so a rare class counts as much as a common one; always guessing one class scores 0.5.
- [Base Rate](https://www.ask-codey.com/glossary/base-rate): How common the positive answer really is in a group, such as the share of its applicants who qualify, before any model is involved.
- [Centroid](https://www.ask-codey.com/glossary/centroid): The center of a cluster: the average of its examples, feature by feature; k-means moves each centroid to its group's average.
- [Chain of Thought](https://www.ask-codey.com/glossary/chain-of-thought): Having a model reason step by step before answering.
- [Class Imbalance](https://www.ask-codey.com/glossary/class-imbalance): When one class is much rarer than another, such as 1 fraud in 100 payments, so accuracy looks high even for a model that never finds it.
- [Clustering](https://www.ask-codey.com/glossary/clustering): Machine learning that finds groups of similar examples on its own, with no answers given, such as customers who visit at the same times.
- [Coefficient](https://www.ask-codey.com/glossary/coefficient): The number a linear model learns for one feature: how much its guess changes for one more unit of it, the other features held the same.
- [Confusion Matrix](https://www.ask-codey.com/glossary/confusion-matrix): A table counting a classifier's right and wrong guesses by kind: true positives, false positives, false negatives and true negatives.
- [Cross-Validation](https://www.ask-codey.com/glossary/cross-validation): Scoring a model several times, each time holding back a different part of the data for testing, then averaging, for a steadier score.
- [Data Drift](https://www.ask-codey.com/glossary/data-drift): When the data a model sees in use changes from what it learned on, such as new kinds of customers, so its guesses may get worse.
- [Data Leakage](https://www.ask-codey.com/glossary/data-leakage): When information the model won't have in real use sneaks into training or testing, so its score looks better than it will be.
- [Decision Path](https://www.ask-codey.com/glossary/decision-path): The questions one example answers on its way down a decision tree, from the top question to the leaf that gives its answer.
- [Decision Threshold](https://www.ask-codey.com/glossary/decision-threshold): The probability at which a classifier switches its answer, 0.5 by default; lower it to catch more, raise it for fewer false alarms.
- [Elbow Method](https://www.ask-codey.com/glossary/elbow-method): Choosing k for k-means by plotting inertia for k = 1, 2, 3 and so on, and picking the k where the drop flattens, like a bent arm.
- [Embedding](https://www.ask-codey.com/glossary/embedding-ai): Numbers capturing text meaning, used for semantic search.
- [Ensemble Learning](https://www.ask-codey.com/glossary/ensemble-learning): Combining several models' guesses, by voting or averaging, so their different mistakes cancel out, as a random forest does.
- [Euclidean Distance](https://www.ask-codey.com/glossary/euclidean-distance): The straight-line distance between two examples, from the Pythagorean theorem: the square root of their squared differences added up.
- [Evaluation (Evals)](https://www.ask-codey.com/glossary/evaluation-evals): Test sets that score AI output quality.
- [Extrapolation](https://www.ask-codey.com/glossary/extrapolation): Guessing for inputs outside the range a model learned from, such as a 110° day when training stopped at 80°; such guesses can't be trusted.
- [F1 Score](https://www.ask-codey.com/glossary/f1-score): One number that combines precision and recall as their harmonic mean, so it's only high when both are.
- [False Positive Rate](https://www.ask-codey.com/glossary/false-positive-rate): Of the real negatives, the share the model wrongly said yes to, FP / (FP + TN), such as the unqualified applicants it invited.
- [Feature Importance](https://www.ask-codey.com/glossary/feature-importance): A score for how much a model relies on each feature; trees' built-in feature_importances\_ and permutation importance are two ways.
- [Feature Scaling](https://www.ask-codey.com/glossary/feature-scaling): Putting features on similar ranges, such as grams and a 1-10 score, so no feature counts more only because its numbers are bigger.
- [Feature Selection](https://www.ask-codey.com/glossary/feature-selection): Choosing which features a model gets, such as the ten most related to the label; done on every row before splitting, it leaks.
- [Feedback Loop](https://www.ask-codey.com/glossary/feedback-loop): When a model's own decisions shape the data its next version learns from, so an early mistake can grow over time.
- [Few-Shot Prompting](https://www.ask-codey.com/glossary/few-shot-prompting): Including examples in the prompt for the model to copy.
- [Grid Search](https://www.ask-codey.com/glossary/grid-search): Trying every combination of a few hyperparameter values, scoring each with cross-validation, and keeping the best.
- [Guardrails](https://www.ask-codey.com/glossary/guardrails): Checks that keep AI inputs and outputs safe.
- [Harmonic Mean](https://www.ask-codey.com/glossary/harmonic-mean): An average that leans toward the smaller numbers, 2 × a × b / (a + b) for two; the F1 score is precision and recall's harmonic mean.
- [Hyperparameter](https://www.ask-codey.com/glossary/hyperparameter): A setting you choose before training, such as a tree's max_depth or the k in k-nearest neighbors, rather than something the model learns.
- [Interpretability](https://www.ask-codey.com/glossary/interpretability): How easily a person can see why a model gave its answer, such as by reading a small tree's questions or a line's coefficients.
- [K-Fold Cross-Validation](https://www.ask-codey.com/glossary/k-fold-cross-validation): Cross-validation that splits the rows into k equal parts and trains k times, each time testing on a different part.
- [K-Means Clustering](https://www.ask-codey.com/glossary/k-means-clustering): Clustering that splits examples into k groups by moving k centers until each example sits with the center nearest to it.
- [k-Nearest Neighbors](https://www.ask-codey.com/glossary/k-nearest-neighbors): A model that guesses by finding the k training examples closest to a new one and taking their most common label, or their average.
- [Linear Regression](https://www.ask-codey.com/glossary/linear-regression): A model that guesses a number with the straight line that runs closest to its examples: a starting point plus a slope times each feature.
- [Logistic Function](https://www.ask-codey.com/glossary/logistic-function): The S-shaped curve that squeezes any number into one between 0 and 1: 0 becomes 0.5, big totals go near 1, very negative ones near 0.
- [Logistic Regression](https://www.ask-codey.com/glossary/logistic-regression): A classification model that turns a weighted sum of the features into a probability between 0 and 1, then picks a class from it.
- [Mean Absolute Error](https://www.ask-codey.com/glossary/mean-absolute-error): The average size of a model's misses, in the label's own units: "off by 8 cups, on average".
- [Mean Squared Error](https://www.ask-codey.com/glossary/mean-squared-error): The average of the squared misses; squaring makes big misses count much more, and the result is in squared units.
- [Missing Indicator](https://www.ask-codey.com/glossary/missing-indicator): An extra 0-or-1 column that says whether a value was missing before it was filled, so the model can learn from the gap itself.
- [Model Card](https://www.ask-codey.com/glossary/model-card): A short document that comes with a model: what it's for, what it learned from, how well it scores for each group, and its limits.
- [Model Persistence](https://www.ask-codey.com/glossary/model-persistence): Saving a trained model to a file so another program can load it and predict without training again.
- [Multiclass Classification](https://www.ask-codey.com/glossary/multiclass-classification): Classification with three or more possible answers, such as apple, orange or pear, where each example gets exactly one.
- [Neural Network](https://www.ask-codey.com/glossary/neural-network): Layers of adjustable weights that learn patterns.
- [One-Hot Encoding](https://www.ask-codey.com/glossary/one-hot-encoding): Turning a category column into one 0-or-1 column per category, such as dog, cat and rabbit columns, so a model can use it.
- [Ordinary Least Squares](https://www.ask-codey.com/glossary/ordinary-least-squares): Fitting a line by choosing the one whose squared errors add up to the least; OLS for short, and what LinearRegression does.
- [Overfitting](https://www.ask-codey.com/glossary/overfitting): When a model learns its examples' random wobble so closely that it does well on them and worse on new examples.
- [Polynomial Features](https://www.ask-codey.com/glossary/polynomial-features): New columns made by raising a feature to powers, such as rainfall² beside rainfall, so a straight-line model can fit a curve.
- [Polynomial Regression](https://www.ask-codey.com/glossary/polynomial-regression): Linear regression on polynomial features, such as rainfall and rainfall², so its guesses follow a curve instead of a straight line.
- [Positive Class](https://www.ask-codey.com/glossary/positive-class): In a yes-or-no classifier, the answer you're looking for, such as fraud or spam; precision and recall measure how well the model finds it.
- [Precision](https://www.ask-codey.com/glossary/precision): Of the examples a model flagged, the share that really were positive: TP / (TP + FP). High precision means few false alarms.
- [Proxy](https://www.ask-codey.com/glossary/proxy-ai): A feature that stands in for another, such as a ZIP code for neighborhood, so a model learns a sensitive feature it was never given.
- [R Squared](https://www.ask-codey.com/glossary/r-squared): How much better a model does than always guessing the average: 1 is perfect, 0 is no better than the average, below 0 is worse.
- [Random Forest](https://www.ask-codey.com/glossary/random-forest): A model made of many decision trees, each trained on a random sample of the rows, that takes their vote or their average.
- [Rate Limits and Token Costs](https://www.ask-codey.com/glossary/rate-limits-and-token-costs): AI APIs charge per token and cap usage per minute.
- [Recall](https://www.ask-codey.com/glossary/recall-ai): Of the real positives, the share the model caught: TP / (TP + FN). Also called sensitivity or the true positive rate.
- [Regression Tree](https://www.ask-codey.com/glossary/regression-tree): A decision tree that guesses a number: each leaf guesses the average label of the training examples that ended there.
- [Reinforcement Learning](https://www.ask-codey.com/glossary/reinforcement-learning): Machine learning where a program learns by trying actions and getting rewards or penalties, such as a game player that improves by playing.
- [Residual](https://www.ask-codey.com/glossary/residual): One example's miss: the actual value minus the model's guess, positive when the guess was too low.
- [Residual Plot](https://www.ask-codey.com/glossary/residual-plot): A chart of each example's residual against the model's guess, so patterns in the misses, such as always too low on big values, stand out.
- [Retrieval-Augmented Generation](https://www.ask-codey.com/glossary/retrieval-augmented-generation): Finding relevant documents and giving them to the model with the question.
- [Root Mean Squared Error](https://www.ask-codey.com/glossary/root-mean-squared-error): The square root of the mean squared error: an average miss in the label's own units that counts big misses more than mean absolute error does.
- [Selection Rate](https://www.ask-codey.com/glossary/selection-rate): The share of a group a model says yes to, such as the invited applicants divided by all applicants from one district.
- [Sensitive Feature](https://www.ask-codey.com/glossary/sensitive-feature): A feature about a personal trait, such as race, sex, age or district, that a model's decisions shouldn't treat people unfairly by.
- [Standardization](https://www.ask-codey.com/glossary/standardization): Rescaling a feature so its mean is 0 and its standard deviation is 1: subtract the mean, then divide by the standard deviation.
- [Stratified Split](https://www.ask-codey.com/glossary/stratified-split): A split that keeps each class's share the same in the training and test parts, such as 20% spam in both; stratify=y does it.
- [Streaming Responses](https://www.ask-codey.com/glossary/streaming-responses): Sending the answer piece by piece as it is generated.
- [Structured Output](https://www.ask-codey.com/glossary/structured-output): Making a model reply in an exact format like JSON.
- [Supervised vs Unsupervised Learning](https://www.ask-codey.com/glossary/supervised-vs-unsupervised-learning): Learning from labelled examples versus finding patterns alone.
- [Tool Use (Function Calling)](https://www.ask-codey.com/glossary/tool-use-function-calling): Letting a model call your functions and use the results.
- [Underfitting](https://www.ask-codey.com/glossary/underfitting): When a model is too simple to catch the pattern, so it does badly on its training examples and on new ones alike.
- [Validation Set](https://www.ask-codey.com/glossary/validation-set): Rows held back to compare models and choose settings, kept apart from the test set, which is saved for one final, honest score.

## Freelancer

- [AI Agent](https://www.ask-codey.com/glossary/ai-agent): A model that plans, uses tools and repeats toward a goal.
- [Bias-Variance Tradeoff](https://www.ask-codey.com/glossary/bias-variance-tradeoff): The balance between a model too simple to catch the pattern (high bias) and one so flexible it learns the noise (high variance).
- [Bootstrap Sample](https://www.ask-codey.com/glossary/bootstrap-sample): A random sample the same size as the data, drawn with replacement, so some rows appear twice and others not at all.
- [Calibration](https://www.ask-codey.com/glossary/calibration): How well a model's probabilities match reality: of the cases it gives 0.7, about 7 in 10 should really be positive.
- [Concept Drift](https://www.ask-codey.com/glossary/concept-drift): When the link between the features and the answer changes, such as what makes a payment fraud, so a model's old lessons stop being true.
- [Demographic Parity](https://www.ask-codey.com/glossary/demographic-parity): A fairness test that asks a model to say yes to the same share of each group, whatever their qualifications.
- [Equalized Odds](https://www.ask-codey.com/glossary/equalized-odds): A fairness test that asks for the same recall (true positive rate) and the same false positive rate in every group.
- [Fine-Tuning](https://www.ask-codey.com/glossary/fine-tuning): Further training a model on your own examples.
- [Gini Impurity](https://www.ask-codey.com/glossary/gini-impurity): How mixed a group's classes are: 0 when all are one class, 0.5 for an even two-class mix; a tree picks the question that lowers it most.
- [Gradient Boosting](https://www.ask-codey.com/glossary/gradient-boosting): An ensemble that builds small trees one after another, each trained to fix the mistakes the trees before it still make.
- [Grounding](https://www.ask-codey.com/glossary/grounding): Tying answers to provided sources.
- [Individual Conditional Expectation](https://www.ask-codey.com/glossary/individual-conditional-expectation): One line per example showing how the model's guess for it changes as one feature changes; their average is the partial dependence plot.
- [Inertia](https://www.ask-codey.com/glossary/inertia): The total of each example's squared distance to its cluster's center; lower means tighter clusters, and it always drops as k rises.
- [Instance-Based Learning](https://www.ask-codey.com/glossary/instance-based-learning): Learning that keeps the training examples themselves and compares each new one to them at guessing time, as k-nearest neighbors does.
- [k-means++](https://www.ask-codey.com/glossary/k-means): A way of choosing k-means' starting centers spread far apart, which makes a poor result less likely; scikit-learn's default.
- [Learning Curve](https://www.ask-codey.com/glossary/learning-curve): A chart of training and validation scores as the training set grows, showing whether more data would help.
- [Local Minimum](https://www.ask-codey.com/glossary/local-minimum): A result that's the best nearby but not the best overall; k-means can settle in one, so it's run from several starts.
- [Model Context Protocol](https://www.ask-codey.com/glossary/model-context-protocol): An open standard for connecting AI to tools and data.
- [Multicollinearity](https://www.ask-codey.com/glossary/multicollinearity): Features that move together, such as a home's size and its bedrooms, so a model can't tell their effects apart and they share the credit.
- [Open-Weight Model](https://www.ask-codey.com/glossary/open-weight-model): A model whose weights are published to run yourself.
- [Partial Dependence Plot](https://www.ask-codey.com/glossary/partial-dependence-plot): A chart of a model's average guess as one feature changes and the rest stay as they are, showing that feature's typical effect.
- [Permutation Importance](https://www.ask-codey.com/glossary/permutation-importance): How much a model's score drops when one feature's values are shuffled; a big drop means the model leans on that feature.
- [Precision-Recall Curve](https://www.ask-codey.com/glossary/precision-recall-curve): A chart of precision against recall at every decision threshold, so you can pick the trade-off between false alarms and misses.
- [Prompt Caching](https://www.ask-codey.com/glossary/prompt-caching): Reusing a repeated prompt prefix to save time and cost.
- [Prompt Injection](https://www.ask-codey.com/glossary/prompt-injection): An attack where input text tries to override the AI's instructions.
- [Regularization](https://www.ask-codey.com/glossary/regularization): A penalty on large coefficients while a model learns, so it can't lean too hard on any one feature; it helps against overfitting.
- [Semantic Search](https://www.ask-codey.com/glossary/semantic-search): Search by meaning rather than exact words.
- [Silhouette Score](https://www.ask-codey.com/glossary/silhouette-score): A score from -1 to 1 for how well examples sit in their own cluster rather than the next nearest; higher means clearer clusters.
- [Target Leakage](https://www.ask-codey.com/glossary/target-leakage): Data leakage through a feature that's only known after the answer, such as a refund column in a fraud model: great test scores, failure in real use.
- [Transformer](https://www.ask-codey.com/glossary/transformer): The network design behind modern LLMs.
- [Validation Curve](https://www.ask-codey.com/glossary/validation-curve): A chart of training and validation scores as one setting changes, such as a tree's depth, showing where underfitting turns into overfitting.

## Master

- [Attention Mechanism](https://www.ask-codey.com/glossary/attention-mechanism): How a model weighs which tokens matter for the next one.
- [Mixture of Experts](https://www.ask-codey.com/glossary/mixture-of-experts): A model routing each token to a few specialist parts.
- [Model Distillation](https://www.ask-codey.com/glossary/model-distillation): Training a small model to copy a larger one.
- [Quantization of Models](https://www.ask-codey.com/glossary/quantization-of-models): Shrinking a model by storing weights with fewer bits.
- [RLHF](https://www.ask-codey.com/glossary/rlhf): Training models on human ratings of their answers.

[More subjects in AI](https://www.ask-codey.com/glossary/groups/ai)

Source: https://www.ask-codey.com/glossary/subjects/ai
