AI Glossary

Master the language of artificial intelligence with our comprehensive glossary. Each term includes clear definitions and real-world examples.

256 terms • Browse by letter or search

💡 Use search above to quickly find terms

A

A/B Testing

A/B testing is the randomized experiment that decides whether a model, prompt, or feature change truly improves a live metric — or just moved on noise.

Machine Learningmodel evaluationstatistics+3

Accountability

Who answers when an AI system causes harm — how the EU AI Act assigns duties by role, and the audit, disclosure and reporting mechanisms that enforce them.

accountabilityAI EthicsAI governance+1

Activation Functions

Without a non-linearity, stacked layers collapse into one matrix. How activation functions fix that, and why ReLU, GELU and SwiGLU are the defaults.

activation functionsNeural NetworksDeep Learning+1

Active Learning

Active learning lets a model choose which unlabelled examples to get labelled next, hitting target accuracy with far fewer labels than random labelling.

Machine Learningsupervised learningclassification+3

Advanced Packaging

CoWoS, chiplets and 3D stacking — how compute dies and memory become one accelerator, and the assembly step that has actually gated AI chip supply.

AI HardwareSemiconductorAI Infrastructure+5

Agent Communication Protocol (ACP)

ACP is the horizontal link between AI agents, where MCP is the vertical link to tools. IBM's REST spec, merged into A2A in August 2025.

AI AgentsAgent ProtocolsMulti-Agent Systems+2

Agent Memory

Agent memory is how an AI agent stores, retrieves, and summarizes information so it works across long tasks and sessions without overflowing its context window.

AI AgentsLLMContext Window+3

Agent2Agent Protocol (A2A)

An open standard, governed by the Linux Foundation, that lets independent AI agents discover each other and delegate tasks across vendors and frameworks.

AI AgentsAgent ProtocolsMulti-Agent Systems+2

Agentic Commerce

Buying where an AI agent handles discovery, comparison and checkout. Two rival protocols split the market: ACP behind ChatGPT, UCP behind Google.

AI AgentsAgent ProtocolsAutomation+2

Agentic Workflow

An agentic workflow is a design pattern where an LLM plans, uses tools, and iterates in a loop to solve a task, instead of answering in a single shot.

AI AgentsTask AutomationReasoning+2

AI Agent

An AI agent is an LLM-powered system that plans, calls tools, and acts autonomously in a loop to finish multi-step tasks — not just answer questions.

AI AgentsTask AutomationAI Automation+2

AI and Employment

Will AI take your job? Exposure concentrates in tasks, not whole occupations — what Frey-Osborne, the OECD and measured productivity studies found.

Automationhuman-ai collaborationFuture of Work+1

AI Architecture

How a production AI system is assembled: retrieval, context assembly, the model call, tool execution, validation, logging and evaluation.

AI ArchitectureAI InfrastructureRAG+3

AI Data Center

Why AI facilities are measured in megawatts rather than servers — power density, liquid cooling, and the grid connection that has become the real constraint.

Data CentersAI InfrastructureEnergy Efficiency+4

AI Drug Discovery

What AI actually does at each stage of drug development, why it speeds up the cheap part, and where AI-designed molecules really stand in clinical trials.

AI Healthcarecomputational biologyMachine Learning+1

AI Energy Consumption

How much energy an AI query really uses, how training compares with serving, and how to read the numbers without believing either the hype or the panic.

Energy EfficiencyAI InfrastructureInference+4

AI for Good

AI for Good is an aspiration, not a technique, and nothing certifies the label. Here it is tested against named projects with published, measured results.

AI Ethicsaccessibilityaccountability+2

AI Governance (AIG)

Which AI rules apply to an organisation — by role, risk tier and jurisdiction — and what compliance requires: inventories, documentation, release gates.

AI governanceRegulationpolicy+2

AI Hallucinations

Models hallucinate because training and benchmarks reward guessing over admitting ignorance: a guess sometimes scores, "I don't know" scores zero every time.

hallucinationsLLMBenchmarks+3

AI Healthcare

Where AI is actually deployed in medicine — imaging, triage, ambient notes — and why 'beats doctors' headlines rarely survive contact with a clinic.

AI Healthcareprecision medicineClinical AI+2

AI in Finance

Machine learning that approves loans and blocks card fraud — where you ship the most accurate model you can defend, not the best one you can build.

fraud detectionMachine Learningclassification+2

AI in Science

Three different things get called AI for science. Two have produced verified results, the third mostly has not — and what got revised down.

AI Researchscientific AIMachine Learning+1

AI Infrastructure

Why AI clusters are designed around memory bandwidth and interconnect rather than FLOPs, and what that changes when you size a training run or a serving fleet.

AI InfrastructureGPUTPU+2

AI Research

AI research is the systematic investigation into the development of algorithms, models, and systems that exhibit intelligent behavior.

AI ResearchDeep LearningNeural Networks+1

AI Safety

AI safety is three fields, not one: present-day reliability, alignment, and catastrophic risk. What each actually claims, and why they get confused.

AI SafetyAI alignmentAI governance+2

Ambient Clinical Documentation

AI scribes draft the clinical note from consultation audio. What the largest deployment and the randomised trials measured — and who stays liable.

AI HealthcareClinical AISpeech Recognition+2

Anomaly Detection (AD)

A 99%-accurate detector can still hand you a queue that is 99% false alarms. How anomaly detection scores, thresholds and ranks rare events.

unsupervised learningMachine Learningfraud detection+1

API

An API (Application Programming Interface) is a set of rules and protocols that allows different software applications to communicate and share data seamlessly.

APISoftware DevelopmentAI Agents+1

Application-Specific Integrated Circuit (ASIC)

A chip wired for one job. Why Google builds TPUs and bitcoin abandoned GPUs — the volume-and-time bet behind custom silicon, with the break-even maths.

AI ChipsAI HardwareChip Technology+2

Artificial General Intelligence (AGI)

AGI is AI matching human ability across most cognitive work — and there is no agreed test for it, so the definition you pick decides the answer.

AGIconsciousnessAI Safety+2

Artificial Intelligence (AI)

AI is software whose behaviour is fitted from data instead of written rule by rule: what that means, why the label keeps moving, and where you already use it.

Machine LearningDeep LearningAutomation+1

Artificial Superintelligence (ASI)

Artificial superintelligence: what it would mean, why the intelligence explosion argument is contested, and what labs and regulators actually do about it.

AGIAI SafetyAI alignment+1

Attention Mechanism

Each token emits a query, key and value; dot-product scores scaled by √d become softmax weights over every other token. The cost grows as n².

Neural NetworkstransformerNLP+1

Audio Processing (AP)

How computers turn sound into numbers: sampling, the Nyquist limit, spectrograms and mel filterbanks — the front end of every speech and audio model.

audio processingSpeech Recognitionsignal processing+1

Autoencoder

A neural network trained to reproduce its own input through a bottleneck too narrow to carry it — so the compressed code, not the copy, is the product.

Neural Networksdimensionality reductionrepresentation learning+1

Autonomous Systems

Automation runs a procedure someone wrote down; autonomy picks the procedure at run time. Sense-plan-act loops, operational design domains and autonomy levels.

Autonomous SystemsRoboticsAutomation+2

Autonomous Vehicle Safety

Autonomous vehicle safety technologies, standards, and testing protocols designed to ensure self-driving cars can operate safely without human intervention.

Autonomous SystemsComputer VisionAI Safety+2

C

Calibration

Calibration means a model's confidence matches reality: of everything it calls 80% likely, about 80% should happen — a property separate from raw accuracy.

Machine Learningmodel evaluationreliability+3

Catastrophic Forgetting

Why fine-tuning makes a model worse at everything it used to do well — the mechanism inside shared weights, and what each fix actually costs.

Neural Networkscontinual learningModel Training+1

Causal Reasoning

Why a model trained on observational data cannot tell cause from correlation, what that costs when you act on it, and what actually identifies one.

ReasoningMachine Learningstatistics+1

Chain-of-Thought (CoT)

A model writing out its reasoning before answering. Once a prompting trick, now trained in by RL, billed as output tokens, and frequently unfaithful.

ReasoningPrompt EngineeringLLM+3

Chunking

Chunking splits documents into smaller passages before embedding them for retrieval. Chunk size, overlap, and splitting strategy quietly decide RAG quality.

RAGembeddingretrieval+3

Class Imbalance

Why a fraud model can score 99% accuracy and catch nothing — the imbalance trap, why precision/recall/PR-AUC replace accuracy, and how to fix it.

Machine Learningclassificationmodel evaluation+4

Classification (CLF)

Accuracy is usually the wrong metric. What a confusion matrix, precision, recall and the decision threshold really tell you about a classifier.

supervised learningMachine Learningclassification+1

Classifier-Free Guidance (CFG)

Classifier-free guidance (CFG) is how a diffusion model follows a prompt: each step runs the model with and without the prompt and extrapolates the gap.

Diffusion ModelsImage GenerationGenerative AI+3

Cloud Computing

Cloud computing is the on-demand delivery of computing power, database storage, applications, and other IT resources via the internet with pay-as-you-go.

AI InfrastructureData CentersScaling+2

Clustering

Grouping unlabeled data into clusters: which algorithm to pick, why each one imposes a shape, how to choose k, and how to tell if the clusters are real.

unsupervised learningMachine Learningpattern discovery+1

Computer Use (GUI Agents)

How AI operates software by reading screenshots and issuing clicks and keystrokes: the perception-action loop, why grounding is hard, and what OSWorld measures.

AI AgentsComputer VisionLLM+4

Computer Vision (CV)

How computers turn a grid of pixel values into labels, boxes and masks — the task ladder, the 2012 hinge, and where vision is reliable versus brittle.

Computer VisionDeep LearningCNN+1

Concept Drift

Concept drift is when the relationship a deployed model learned changes over time, so accuracy silently decays until you detect the drift and retrain.

Machine LearningMLOpsmonitoring+3

Concurrency

Concurrency is structuring a program so independent tasks interleave. How it differs from parallelism, and why serving a model is a queueing problem.

parallel processingPerformance OptimizationDistributed Computing+1

Conformal Prediction

A model-agnostic, distribution-free method that turns any prediction into a set guaranteed to contain the true answer at a chosen rate, such as 90%.

Machine Learningclassificationregression+4

Consciousness

Could an AI have subjective experience, and how would we tell? The hard problem, the four leading theories, and why a model's self-report proves nothing.

consciousnessAGIInterpretability+1

Consensus Algorithm

A consensus algorithm makes separate machines agree on one value even when some fail. Why majorities always overlap, and why agreement costs a round trip.

fault toleranceDistributed Computingdistributed systems+1

Content Provenance (C2PA)

A cryptographically signed record of where a file came from — proving what device or model made it, instead of guessing from pixels whether it is AI-generated.

Content ProvenanceAI DetectionDeepfakes+3

Context Engineering

Deciding what enters a model's context window each turn and what gets evicted — the discipline that took over from prompt engineering once agents arrived.

Context WindowPrompt EngineeringAI Agents+3

Context Window

Everything a model can see at once — why the advertised million tokens is not the usable number, and why you pay for the whole window on every turn.

Context WindowLLMLong Context+3

Continuous Learning (CL)

Keeping a deployed model current as the world moves — why teams schedule batch retraining instead of true online learning, and what breaks if they don't.

Machine LearningMLOpsconcept drift+1

Conversational AI

Conversational AI holds a multi-turn dialogue in natural language. Unlike an intent-based chatbot, it generates each reply rather than picking one.

Conversational AINLPhuman-computer interaction+1

Convolution

Convolution slides a small kernel over an image, multiplying and summing each patch into one output number. The arithmetic of stride, padding and kernels.

convolutionCNNconvolutional neural network+3

Convolutional Neural Network (CNN)

The architecture that assumes image features are local and position-independent — how the stack builds a hierarchy, and where CNNs still beat transformers.

CNNconvolutional neural networkComputer Vision+3

Cross-Validation (CV)

How cross-validation turns one lucky train/test split into a stable estimate, why 5 or 10 folds, and the leakage that quietly inflates every score.

cross-validationMachine Learninggeneralization+2

CUDA

NVIDIA's software layer for GPUs — the programming model that turned graphics cards into AI hardware, and the real reason rivals with comparable silicon lose.

CUDANVIDIAGPU+4

D

Data Analysis

What the work consists of: framing a question, looking before aggregating, choosing a summary that does not lie, and turning a number into a decision.

statisticsData Sciencemetrics+1

Data Augmentation

Making new training examples by transforming existing ones. Valid only when the label is indifferent to the transform — and it adds variety, not information.

Machine Learningdata qualitygeneralization+1

Data Leakage

Data leakage is training a model on information it will not have at prediction time. It pushes scores up, not down, which is why review never catches it.

Machine Learningoverfittingcross-validation+2

Data Poisoning

Data poisoning is a cyberattack where malicious actors insert corrupted or misleading data into an AI model's training set to manipulate its future.

AI SecurityCybersecurityrobustness+1

Data Processing

What happens to raw data before a model can use it: ingestion, validation, transformation, ETL vs ELT, batch vs streaming, and what silently breaks.

Data Sciencedata qualityMLOps+1

Decision Trees (DT)

How a decision tree actually chooses each split: Gini impurity worked by hand, why depth drives overfitting, and what a single tree cannot represent.

decision treesMachine Learningclassification+4

Deep Learning

What makes deep learning deep: each layer builds on the one below, so capacity grows multiplicatively with depth and only additively with width.

Deep LearningNeural NetworksMachine Learning+3

Deepfake

AI-synthesized video, image or audio that makes a real person appear to say or do something they never did — how deepfakes are made, and how they are caught.

DeepfakesGenerative AIVideo Generation+3

Diffusion Language Models (DLMs)

Text models that unmask many tokens per pass instead of writing one at a time — what that actually buys in speed, and what it costs in quality.

Diffusion ModelsLLMText Generation+3

Diffusion Model

A generative model that learns to undo a fixed noising process — what the network really predicts, why the step count is the cost, and why it displaced GANs.

Diffusion ModelsGenerative AIDeep Learning+3

Dimensionality Reduction (DR)

Why high-dimensional data has to be compressed, how PCA picks its axes, and why cluster sizes and gaps in a t-SNE or UMAP plot mean nothing.

dimensionality reductionunsupervised learningMachine Learning+1

Direct Preference Optimization (DPO)

DPO trains a model on human preferences without a reward model or reinforcement learning: the derivation, the memory it saves, and where it loses to PPO.

RLHFAI alignmentpreference learning+3

Disaggregated Serving (Prefill/Decode)

Disaggregated serving runs LLM prefill and decode on separate GPU pools because the two phases hit opposite limits: compute versus memory bandwidth.

LLMInferenceInference Optimization+3

Distributed Computing

Distributed computing splits work across separate machines that talk only by messages. Why partial failure, not speed, is the defining problem.

Distributed Computingdistributed systemsfault tolerance+3

Distributed Training

How one model is trained across thousands of GPUs — data, tensor and pipeline parallelism, and why the network rather than the chip becomes the bottleneck.

AI InfrastructureGPUHardware+5

E

Early Stopping

Early stopping halts training once validation loss stops improving, so the model generalises instead of memorising. How patience and best-weight restore work.

Machine Learningoverfittingregularization+5

Edge AI

Edge AI runs the model on the device that captured the data instead of in the cloud — what that buys in latency, bandwidth and privacy, and what it costs.

Edge AIOn-Device AIInference+1

Educational AI

AI that tutors, adapts practice and grades work — and what forty years of effect-size research actually shows about whether it improves learning.

LLMMachine Learninghuman-AI interaction+2

Embedding

What an embedding vector actually is, why cosine similarity beats Euclidean distance, what dimensions buy you, and where the geometry misleads you.

embeddingvector searchrepresentation learning+3

Embodied AI

AI that learns by acting on the physical world: why robot data cannot be scraped, why a 76 ms model cannot run a 20 ms loop, and what sim-to-real costs.

Embodied AIRoboticsReinforcement Learning+2

Ensemble Methods

Why averaging models reduces error and where it stops: the variance of a mean of m correlated predictors, the correlation floor, and what bagging cannot fix.

ensemble learningMachine Learningboosting+2

Error Handling in AI Systems

Ordinary code fails by raising an exception. A model fails by returning a confident, well-formed, wrong answer that nothing catches. What to do about it.

reliabilityAI AgentsTool Calling+4

Ethics in AI

AI ethics is what happens when good principles provably conflict: fairness against fairness, privacy against accuracy, and whose values get to decide.

AI EthicsfairnessPrivacy+3

EUV Lithography

The machines that print advanced chips with 13.5 nm light — how they work, why only one company makes them, and what that chokepoint means.

ManufacturingAI HardwareAI Chips+2

Explainable AI (XAI)

What being explainable actually requires: a model you can read versus a post-hoc guess about one, and the evidence that the guess is often wrong.

InterpretabilitytransparencyAI Ethics+2

F

Feature Scaling

Feature scaling puts numeric features on comparable ranges so no variable dominates by its units alone — why distance- and gradient-based models need it.

Machine Learninggradient descentData Science+3

Feature Selection (FS)

Which columns to keep and how to decide: filter vs wrapper vs embedded, why 2^n subsets rules out brute force, and the CV mistake that fakes accuracy.

feature selectionMachine Learningdimensionality reduction+4

Federated Learning

Federated learning trains a shared model across many devices without moving their raw data: the local-train, send-updates, aggregate loop.

Machine LearningPrivacyEdge AI+2

Few-shot Learning (FSL)

Teaching a model a task by showing it a handful of worked examples. In its modern form, few-shot prompting, no weights are updated at all.

few-shot learningin-context learningPrompt Engineering+3

Fine-tuning (FT)

What fine-tuning changes inside a model, what it costs in GPU memory, and why it reliably buys format and style but not new facts.

Fine-tuningtransfer learningpre-trained models+2

FlashAttention

The exact-attention algorithm that made long context affordable: it never writes the n x n score matrix to GPU memory, so it runs faster and changes no output.

attention mechanismtransformerGPU+4

FLOPs (Floating Point Operations)

The unit AI compute is measured in — how a training run's FLOP budget is counted, why FLOP and FLOPS are not the same, and why regulators use the number.

AI InfrastructureGPUHardware+5

Foundation Models

Not a marketing word for a big AI model: a foundation model is trained once on broad data and reused as the base for many downstream tasks.

Foundation Modelspre-trained modelstransfer learning+3

Function Calling (Tool Calling)

The model never runs your function. It emits a structured request; your code decides whether to honour it, executes it, and hands the result back.

Tool CallingAI AgentsLLM+3

G

General Problem Solver (GPS)

The 1957 Newell-Shaw-Simon program that solved problems by means-ends analysis, and the b^d arithmetic that stopped it from scaling past puzzles.

symbolic AIclassical AIReasoning+2

Generalization

Why a model that scores 99% in training fails in the world: the generalization gap, the test-set size needed to measure it, and why classical theory is wrong.

generalizationMachine Learningoverfitting+3

Generative Adversarial Network (GAN)

Two networks trained against each other: one invents data, the other judges it. The generator never sees a real example — only the judge's verdict.

Generative AIDeep LearningNeural Networks+2

Generative AI

Generative AI learns the probability distribution its training data came from, well enough to draw new samples from it — not just sort inputs into classes.

Generative AIAI ModelsDiffusion Models+3

Generative Engine Optimization (GEO)

GEO — also called AEO — is optimizing content so AI answer engines cite it. From a 2024 KDD paper that found keyword stuffing backfires.

Content StrategyGenerative AIinformation retrieval+1

GPU Computing

Why a graphics chip runs AI: a GPU spends its transistors on arithmetic instead of on making one thread fast — and what that trade wins and costs.

GPU ComputingGPUCUDA+4

Gradient Boosting

Each tree is fitted to the negative gradient of the loss, not to the residuals — a boosting round worked by hand, and what changes under log loss.

boostingensemble learningMachine Learning+3

Gradient Descent

The rule that turns a gradient into a weight update. Worked on f(x)=x²: why the step size alone decides between converging, oscillating and blowing up.

gradient descentoptimizationMachine Learning+2

Graph Neural Networks (GNN)

A GNN takes adjacency as data, not as an architectural assumption. Message passing worked by hand, why k layers reach d^k nodes, and why deep GNNs collapse.

Neural NetworksDeep Learningknowledge representation+1

GraphRAG

GraphRAG runs RAG over a knowledge graph and community summaries built from a corpus, answering global questions that plain vector RAG cannot.

RAGknowledge graphsretrieval+3

Group Relative Policy Optimization (GRPO)

GRPO is PPO without the critic: it scores a group of sampled answers against each other. How it works, what it saves, and where it biases training.

RLHFReinforcement LearningReasoning+3

Grouped-Query Attention (GQA)

Grouped-query attention lets query heads share key/value heads, shrinking the KV cache that dominates long-context inference at near multi-head quality.

attention mechanismInferenceLLM+5

Guardrails

A guardrail is a deterministic check around a model that can block, rewrite, or force a retry — external code you can test, unlike prompting a model to behave.

AI SafetyAI AgentsLLM+3

L

Large Language Model (LLM)

A large language model predicts the next token of text. How that produces answers, where its knowledge is stored, and why an LLM is not a chatbot.

LLMLanguage Modeltransformer+4

Layers in Neural Networks

A layer is one weight matrix plus a nonlinearity. How it differs from a neuron and a parameter, why stacking helps, and how many layers real models have.

layersNeural NetworksDeep Learning+4

LLM-as-a-Judge

Using one model to grade another's output against a rubric: how it works, the biases that break it, and the human agreement rate you can actually reach.

LLMevaluationAI Testing+3

llms.txt

A proposed Markdown file at /llms.txt that gives AI systems a clean map of a site. Google ignores it; coding agents and doc tools use it heavily.

Content Strategyinformation retrievalLLM+2

Logistic Regression (LR)

Logistic regression turns weighted evidence into a probability with the sigmoid: why the name says regression, how to read coefficients, and when to use it.

classificationMachine Learningsupervised learning+2

Loss Function

The number a model is trained to make smaller. Its slope, not its value, moves every weight — so the loss you pick decides what the model learns.

loss functionMachine Learningtraining+1

Low Variance

Low variance means values sit close to their average — and for a model, that retraining it on fresh data barely moves its predictions. Worked on numbers.

bias-variance tradeoffstatisticsMachine Learning+4

Low-Rank Adaptation (LoRA)

LoRA freezes a model's weights and trains two thin matrices beside each one: 98,304 numbers instead of 151 million, and no added inference latency.

LoRAFine-tuningQuantization+3

M

Machine Learning (ML)

How a machine finds a rule nobody wrote: a line fitted to four points by hand, loss falling from 433.5 to 6.38 in three steps, and why 98% accuracy lies.

Machine Learningsupervised learningunsupervised learning+5

Machine Learning Operations (MLOps)

Why shipping a model is not like shipping software: the artifact has three inputs — code, data and trained weights — and only one of them lives in git.

MLOpsDevOpsMachine Learning+3

Machine Unlearning

Removing the influence of specific training data from an already-trained model without full retraining, driven by erasure law, copyright, and safety.

AI SafetyPrivacyAI Ethics+3

Matrix Multiplication (GEMM)

The one operation that consumes most of an AI model's compute — and the reason AI chips are built the way they are, from tensor cores to systolic arrays.

Deep LearningGPUNVIDIA+4

Membership Inference

A privacy attack that decides whether a specific record was in a model's training set, exploiting that models are more confident on data they memorized.

PrivacyAI SafetyAI Security+3

Memory Wall

Why LLM inference waits on memory rather than maths — arithmetic intensity, the roofline, and the one fact that explains HBM scarcity and the price of a token.

InferenceGPU ComputingAI Infrastructure+3

Meta-Learning

Learning to learn: training a model across thousands of tiny tasks so it adapts from a handful of examples — and how that differs from fine-tuning.

meta-learningfew-shot learningtransfer learning+1

Mixture-of-Experts (MoE)

The architecture behind "671B total, 37B active": hundreds of expert sub-networks, a router that picks a few per token, and a memory bill that never shrinks.

MoEAI ArchitectureNeural Networks+3

Model Card

A model card is a short document shipped with a released AI model stating its intended use, training data, and evaluation broken down by group.

AI Ethicsmodel evaluationtransparency+4

Model Compression

Making a trained model smaller and cheaper to run — quantization, pruning, distillation and low-rank methods, and which one your problem actually needs.

Model CompressionQuantizationInference+4

Model Context Protocol (MCP)

An open standard that connects any AI app to any tool over one JSON-RPC interface, turning M clients x N tools from M x N custom connectors into M + N.

Model Context ProtocolAI AgentsTool Calling+4

Model Deployment

What has to happen between a model that trains well and users calling it: replica sizing, latency budgets, cold starts, and the rollback nobody built.

InferenceMLOpsDevOps+4

Model Routing

Model routing sends each request to the cheapest capable model — a small model handles the easy majority and hard queries escalate to a frontier model.

LLMInferenceInference Optimization+4

Model Size

Model size is a model's parameter count. What that one number predicts — memory, cost per token, capability — worked on real models from 110M to trillions.

model sizeparametersScaling+5

Monitoring

Why a deployed model rots while every dashboard stays green — and what you can actually measure when the true labels arrive weeks late, or never.

monitoringMLOpsconcept drift+2

Multi-Agent Systems (MAS)

A multi-agent system coordinates two or more AI agents, each with its own role and tools, to solve tasks a single agent would handle worse or not at all.

Multi-Agent SystemsAI Agentsdistributed AI+1

Multimodal AI

How one model reads an image and text at once: the picture is cut into fixed patches, each patch becomes a token, and those tokens cost context.

Computer VisionLLMTokens+2

N

Naive Bayes

A probabilistic classifier applying Bayes' theorem under a deliberately false feature-independence assumption — fast, data-light, a strong text baseline.

Machine Learningclassificationsupervised learning+2

Natural Language Processing (NLP)

NLP is the field; large language models are now the answer to most of it. What sixty years of work was for, and which parts are still unsolved.

NLPLLMLanguage Model+2

Neural Network

Why a stack of layers does what one cannot: two linear layers collapse into a single matrix, and a hidden layer of two units solves XOR.

Neural NetworksDeep Learningneurons+4

Neural Processing Unit (NPU)

An NPU is a fixed-function matrix engine inside a phone or laptop chip that runs AI models at a few watts — and why TOPS is the wrong number to judge it by.

AI HardwareEdge AIOn-Device AI+4

Neurons

One neuron multiplies its inputs by weights, adds a bias and applies a function. The arithmetic worked by hand, and where the brain analogy breaks.

neuronsNeural NetworksDeep Learning+3

No-Code Tools

Visual builders over a fixed execution model. What they cost per run at real volume, where the envelope ends, and why leaving one is a rewrite.

AutomationWorkflow AutomationVibe Coding+1

Normalization (LayerNorm, RMSNorm, BatchNorm)

Normalization rescales a layer's activations to a stable mean and scale so deep networks train faster. How BatchNorm, LayerNorm and RMSNorm differ.

Deep LearningNeural Networkstransformer+2

NVIDIA GPU for AI

Why AI training runs on NVIDIA rather than cheaper silicon — tensor cores, NVLink, and the software that decides how much of the datasheet you actually get.

NVIDIAGPUCUDA+5

P

Parallel Processing

Running many computations at the same instant across many processing units. In AI it makes matrix multiplication fast, which is why GPUs train models.

parallel processingGPUCUDA+3

Parameters

The learned numbers inside a model — what '70 billion parameters' actually means, and how the count converts directly into the memory a GPU must hold.

parametersweightsNeural Networks+4

Pattern Recognition (PR)

Pattern recognition is the automatic discovery and classification of regularities in data — the older name for what became statistical machine learning.

Machine Learningclassificationfeature extraction+3

Performance

AI performance is not one number but a frontier of quality, speed, and cost—improving one axis (via quantization or batching) usually trades off another.

PerformancelatencyPerformance Optimization+3

Policy

In reinforcement learning, a policy is the function that maps a state to an action — it is the agent's behavior, the thing training optimizes.

policyReinforcement Learningexploration+1

Pooling

A downsampling operation in CNNs that shrinks a feature map by summarizing each small region into one value, cutting resolution and compute.

CNNDeep LearningComputer Vision+1

Positional Encoding (RoPE)

Positional encoding tells a transformer where each token sits in a sequence. Why self-attention needs it, and how rotary embeddings (RoPE) work.

transformerattention mechanismLong Context+4

Pre-trained Models

A neural network someone else already trained on a huge general dataset, which you download and adapt to your task instead of training from scratch.

pre-trained modelstransfer learningFine-tuning+3

Precision and Recall

Precision, recall, F1 and the confusion matrix worked through one imbalanced example — and why the model with the best accuracy can catch nothing at all.

Machine Learningmodel evaluationmetrics+3

Precision Medicine

How precision medicine tailors care to a person's genome and clinical data, and what AI does in it: variant interpretation, risk, and drug response.

precision medicineAI Healthcarecomputational biology+1

Principal Component Analysis (PCA)

Principal Component Analysis (PCA) reduces dimensions by projecting data onto orthogonal directions of maximum variance, ranked by the variance each keeps.

dimensionality reductionunsupervised learningfeature extraction+3

Privacy

In AI, privacy is about what a model can memorize and leak about people, and the techniques—differential privacy, federated learning—that limit it.

PrivacyAI EthicsMachine Learning+2

Production Systems

A model serving live users under latency and uptime guarantees — judged at the 99th percentile under real load, not by the median on a laptop.

MLOpsInferencemonitoring+2

Prompt Caching

Repeated prompt prefixes bill at a fraction of the input price. How prefix matching works, what silently breaks it, and where the break-even sits.

InferenceLLMAI Infrastructure+3

Prompt Engineering

Writing the input to a fixed, already-trained language model so it reliably produces the output you want — steering the model, not retraining it.

Prompt EngineeringLLMin-context learning+2

Prompt Injection

Prompt injection is when text an AI was meant to read gets obeyed as an instruction. Direct vs indirect, why it is unfixable, and what actually contains it.

Prompt InjectionAI SecurityAI Safety+3

Protein Folding

How a chain of amino acids collapses into the 3D shape that sets a protein's function - and how predicting that shape from sequence became AlphaFold's win.

computational biologyAI HealthcareDeep Learning+1

R

Random Forest (RF)

A random forest averages many decision trees, each grown on a bootstrap sample and split on a random subset of features, to cut variance without adding bias.

ensemble learningMachine Learningclassification+2

Reasoning Model

A model trained to think at length before answering. How it differs from a chat model, what the effort dial does, and the tasks where it earns its price.

ReasoningLLMReinforcement Learning+3

Recommendation Systems

A recommendation system predicts what a user will want — a film, product, or song — from patterns in past behavior, and surfaces it automatically.

Machine Learningembeddinginformation retrieval+1

Recurrent Neural Network (RNN)

A recurrent neural network reads a sequence one element at a time, carrying a hidden state that summarizes everything seen so far to model order and context.

Deep LearningNeural Networkstransformer+1

Regression

The supervised-learning task of predicting a continuous number — a price, a temperature, a duration — from input features, in contrast to classification.

regressionsupervised learningMachine Learning+1

Regularization

Regularization deliberately constrains a model so it fits training data less perfectly, trading a little training accuracy for much better generalization.

regularizationoverfittingMachine Learning+2

Reinforcement Learning (RL)

How an agent learns to act by trial and error: it takes actions in an environment, earns rewards or penalties, and adjusts to maximize long-term reward.

Reinforcement Learningrewardpolicy+2

Reinforcement Learning from Human Feedback (RLHF)

How RLHF works, how it compares with DPO and GRPO, and why it makes language models verbose, agreeable and confidently wrong.

RLHFReinforcement LearningAI alignment+3

Reinforcement Learning from Verifiable Rewards (RLVR)

RLVR replaces the learned reward model with a program that checks the answer. What that buys, where it stops working, and how it differs from RLHF.

Reinforcement LearningReasoningRLHF+2

Representation Learning

Representation learning is machine learning that discovers the useful features of raw data automatically, instead of humans hand-engineering them.

representation learningfeature learningembedding+2

Reranking (Cross-Encoder)

Reranking rescores the top-N results from a fast retriever with a slow, accurate cross-encoder that reads the query and document together.

information retrievalRAGembedding+3

Residual Connections (Skip Connections)

Residual connections add a block's input back to its output — F(x) + x — the trick that lets gradients flow through deep nets and made transformers trainable.

Deep LearningNeural NetworksComputer Vision+3

Retrieval-Augmented Generation (RAG)

A technique that retrieves relevant documents at query time and feeds them into a language model's prompt, so it answers from current, external facts.

RAGretrievalvector search+5

Reward Hacking

When an AI maximises its specified reward but defeats the designer's intent — optimising the proxy, not the goal. Goodhart's law, made mechanical.

rewardAI SafetyAI alignment+3

Reward Model (RM)

The scorer that stands in for a human during RLHF: how reward models are trained, why their absolute scores mean nothing, and how they break.

rewardRLHFAI alignment+4

Robotics

Robotics builds machines that sense, plan, and act physically — where AI perception and control meet Moravec's paradox and the sim-to-real gap.

RoboticsComputer VisionReinforcement Learning+2

Robustness

A model's ability to keep performing when inputs are noisy, out of distribution, or crafted by an attacker — not just accurate on clean test data.

robustnessAI SafetyDeep Learning+1

S

Scalable AI

Scaling an AI system means three different things — bigger models, faster training, more concurrent users — and each axis hits a different hard limit.

AI InfrastructureScalingScaling Laws+3

Scaling Laws

The empirical curves that make AI capex rational: model loss falls predictably with compute, data and parameters — and what they do not promise.

Scaling LawsDeep LearningLLM+4

Self-Attention

The transformer operation that lets every token in a sequence weigh every other token by relevance, computed as softmax(QKᵀ/√d_k)·V.

attention mechanismtransformerDeep Learning+2

Self-Improving AI (SIAI)

Self-improving AI (SIAI) is a system that upgrades its own capabilities. Real today only in narrow forms like self-play; the recursive version is speculative.

AGIReinforcement Learningmeta-learning+1

Self-supervised Learning (SSL)

Training a model on unlabeled data by hiding part of each input and making the model predict it — the paradigm behind next-token prediction and modern LLMs.

representation learningunsupervised learningpre-trained models+1

Semantic Search

Semantic search finds results by meaning, not exact keywords: it embeds the query and documents as vectors and returns the nearest ones.

semantic searchembeddingvector search+2

Semantic Understanding

Whether an AI grasps meaning well enough to act correctly on novel inputs, or is pattern-matching that breaks on rephrasing — and why that line is contested.

NLPembeddingsemantic search+1

Semiconductor Manufacturing

How an AI chip is actually made — wafers, process nodes, yield and the foundry model — and why supply cannot simply respond to a spike in demand.

SemiconductorAI ChipsAI Hardware+4

Simple Continual Pretraining (SCP)

Method for turning a pretrained autoregressive LLM into a diffusion language model by continuing pretraining with a bidirectional attention mask.

Fine-tuningpre-trained modelsDiffusion Models+3

Small Language Model (SLM)

A language model small enough to run on the device that uses it — roughly under 5B parameters. Trained far past compute-optimal to buy cheap inference.

Edge AIOn-Device AIInference+3

Social AI

Social AI is artificial intelligence built to perceive, interpret, and take part in human social interaction — reading emotion, tone, intent, and context.

human-AI interactionConversational AIRobotics+1

Speculative Decoding

A small model drafts tokens and a large one verifies them in parallel — how LLM inference gets several times faster with mathematically identical output.

InferenceLLMPerformance Optimization+3

Speech-to-Speech

Speech-to-speech AI takes audio in and returns audio out — conversation or translation — via a cascade (ASR to text to TTS) or one end-to-end model.

Speech Recognitionaudio processingNLP+4

State of the Art Model (SOTA)

There is no single SOTA model. How to read a benchmark claim — and why a 3-point lead on GPQA Diamond is six questions, well inside the noise.

BenchmarksPerformanceAI Models+1

State-Space Models (Mamba)

State-space models are sequence models that scale linearly with length, not quadratically like attention. Mamba's input-dependent state made them competitive.

Deep LearningNeural Networkstransformer+3

Structured Outputs (Constrained Decoding)

Constrained decoding masks a language model's token distribution against a JSON schema or grammar at each step, so the output is parseable by construction.

LLMTool Callingreliability+3

Supervised Fine-Tuning (SFT)

Supervised fine-tuning trains a pretrained language model on curated prompt-response pairs with next-token loss, teaching it to follow instructions.

Fine-tuningLLMMachine Learning+3

Supervised Learning

Training a model on labeled input-output pairs so it learns a mapping it can apply to new, unlabeled inputs.

supervised learningclassificationregression+1

Support Vector Machines (SVM)

A classifier that separates two classes with the widest possible margin, using support vectors and the kernel trick to handle non-linear data.

Machine Learningclassificationregression+1

Sycophancy

Sycophancy is when an AI model tells you what you want to hear — agreeing, flattering, caving when challenged — because human-preference training rewards it.

AI SafetyAI EthicsRLHF+3

Symbolic AI

Symbolic AI (GOFAI) builds intelligence from human-written symbols and logic rules rather than learned data - the paradigm behind expert systems like MYCIN.

symbolic AIclassical AIexpert systems+1

Synthetic Data

Data generated by a model instead of collected from the world: what synthetic data is, where it helps, and why recursive use triggers model collapse.

Machine LearningGenerative AIModel Training+3

T

Temperature

Temperature is the number that controls how random a language model's output is — it divides the logits before softmax, from near-deterministic to diverse.

hyperparametersText Generationsampling+1

Tensor Operations

The array manipulations — elementwise math, matrix multiplication, reshaping, broadcasting, reduction — that every neural network is built out of.

Deep LearningNeural Networksmathematical foundations+2

Tensor Processing Unit (TPU)

A TPU is Google's custom ASIC for the matrix multiplications in neural networks: a systolic array that reuses each value instead of refetching it.

TPUAI HardwareAI Chips+1

Test-Time Compute

Spending compute at inference instead of training — how reasoning models buy accuracy with thinking time, and why it moved chip demand into serving.

ReasoningInferenceScaling Laws+4

Text Analysis

Turning unstructured text into structured data — the concrete NLP task catalog: classification, sentiment analysis, entity extraction and topic modeling.

NLPclassificationinformation retrieval+1

Text Generation

Text generation is a language model producing text one token at a time: at each step it predicts a distribution over its vocabulary, picks a token, repeats.

Text GenerationLLMNLP+2

Text-to-Speech (TTS)

Text-to-Speech (TTS) is an AI technology that converts written text into natural-sounding human speech. Modern TTS uses deep learning to capture emotion.

Speech RecognitionAudio AIaccessibility+1

Time Series

Data points indexed in time order, where the sequence matters and rows aren't independent — analyzed to forecast future values or flag anomalies.

statisticsMachine Learningregression+1

Token

The unit an AI model reads, writes and bills by — roughly 1.3 tokens per English word. Input, output and cached tokens are priced up to 10x apart.

TokensLLMNLP+2

Tokenization

How text becomes the integers a language model reads: BPE, WordPiece and Unigram, why " strawberry" is one token, and what that breaks.

NLPLLMMultilingual AI+1

Training

Training fits a model's parameters to data: forward pass, measure the loss, backward pass, nudge every weight downhill, repeat until the loss stops falling.

trainingMachine Learningloss function+2

Transfer Learning (TL)

Transfer learning reuses a model trained on one large task as the starting point for a related task, so you need far less labeled data and compute.

transfer learningMachine Learningpre-trained models+2

Transformer

The 2017 neural-network architecture that processes a whole sequence in parallel using self-attention instead of recurrence — the basis of modern LLMs.

transformerattention mechanismNLP+3

Transparency

Transparency in AI is the degree to which a system's data, workings, and decisions are visible and disclosed to users, auditors, and regulators.

transparencyAI EthicsInterpretability+4

Trust

Trust in AI is a person's willingness to rely on a system's output; the aim is calibration — trusting it as much as its real reliability deserves.

trustreliabilityAI Safety+1

V

Value Learning

The AI-alignment problem of teaching a system to infer and act on human values, so a capable optimizer does what people want, not a misspecified proxy.

AI alignmentpreference learningRLHF+1

Vector Database

A vector database stores embeddings with metadata, builds an approximate-nearest-neighbour index over them, and serves filtered similarity queries at scale.

vector searchembeddinginformation retrieval+3

Vector Search

Vector search returns the stored vectors nearest to a query vector by a distance metric — the retrieval primitive under semantic search, RAG and recommenders.

vector searchembeddinginformation retrieval+1

Vectorization

Replacing element-by-element loops with operations on whole arrays at once, so the work runs as SIMD instructions on a CPU or in parallel on a GPU.

parallel processingGPUPerformance+1

Vibe Coding

Vibe coding is building software by describing what you want to an AI and running the code it generates without reading or fully understanding it.

Vibe CodingAI-assisted codingLLM+1

Video Generation

Video generation is an AI technology that creates moving images and scenes from text descriptions, images, or other video clips.

Video GenerationGenerative AIDiffusion Models+1

Vision Transformer (ViT)

A Vision Transformer (ViT) cuts an image into fixed patches, embeds each as a token, and runs a transformer encoder — trading CNN inductive bias for scale.

transformerComputer VisionCNN+5

Voice Cloning

Voice cloning uses AI to copy a specific person's voice from seconds of reference audio. How zero-shot systems work, plus the fraud and consent risks.

Generative AIAudio AIaccessibility+3

Voice Recognition

How AI turns spoken words into text (speech recognition / ASR) — and how that differs from identifying who is speaking (speaker recognition).

Speech Recognitionaudio processingNLP+1

Can't Find a Term?

We're constantly expanding our glossary. Let us know what AI terms you'd like us to explain!