Skip to content

The AI glossary

AI, in plain English.

People throw AI terms around as though we all got the same briefing. We didn't. Here's what they mean, where they show up, and where I'd go to learn more.

192 terms · Curated by Matthew Williamson · Updated

New here? Start withAILLMsAgentsHallucinations

A29 terms

Under the hood

Abductive logic programming

ALP

A logic-programming framework that searches for assumptions which, together with existing rules, explain an observation or satisfy a goal. The proposed assumptions must also obey specified integrity constraints.

Under the hood

Ablation

Ablation study · Ablation studies

Removing, disabling, or changing part of a model or system and comparing the results. Researchers use ablations to investigate the effect of components such as attention heads, input features, tools, or training choices.

Risks & judgment

Abstention

Selective prediction · Selective answering

When an AI system leaves a question unanswered because it lacks enough evidence or confidence. It might say it doesn't know, request more information, or pass the question to a person.

Risks & judgment

Adversarial

Adversarial attack · Adversarial testing · Adversarial examples · Adversarial machine learning

Designed to challenge or work against a system. In AI, adversarial attacks deliberately manipulate inputs, training data, or other parts of a system to cause mistakes or unwanted behavior. Adversarial testing probes weaknesses so they can be understood and addressed.

Agents & tools

Agent dreaming

Sleep-time compute · Background memory consolidation · Dream · Dreaming

An informal name for background processing that reviews an agent's experiences and reorganizes stored information for future use. Some systems summarize conversations, connect related notes, or prepare useful context between interactions. Sleep-time compute is one approach.

Agents & tools

Agent handoff

Handoff · Agent handover

Passing responsibility for a task or conversation from one agent to another, along with the context needed to continue. A handoff changes who handles the next part of the work.

Agents & tools

Agent identity

AI agent identity · Workload identity

The identifier and authentication mechanisms that let other systems recognize an agent as a distinct software actor. An identity can be associated with permissions, ownership, and records of actions.

Agents & tools

Agent loop

Agentic loop

The repeated cycle in which an agent receives information, chooses a next step, takes an action, and uses the result to decide what follows. The loop ends when the task is finished, a limit is reached, or the system pauses for help.

Agents & tools

Agent registry

AI agent registry · Agent directory · ERC-8004

A directory that lets people and software discover agents and look up information about them, such as identifiers, service addresses, capabilities, or feedback. Registries can use ordinary databases or blockchains. The draft ERC-8004 proposal describes blockchain registries for agent identity, reputation, and validation.

Agents & tools

Agent skills

Agent skill · AI skills · SKILL.md

Reusable packages of instructions and resources that help an agent carry out particular tasks. A skill might explain how to edit a spreadsheet, include a script to check it, and supply examples of the expected result.

Agents & tools

Agent2Agent protocol

A2A · Agent-to-agent protocol

An open protocol for independent agent applications to discover capabilities, exchange messages, and coordinate tasks. It defines ways to describe agents and track work across systems that may use different models or frameworks.

Agents & tools

Agentic coding

Agentic programming

Using AI agents to carry out software development tasks across multiple steps. An agent might inspect a repository, edit files, run tests, and revise its changes based on the results.

Agents & tools

AI agent

Agentic AI · Agents

An AI system that can choose steps, use tools, and respond to what happens as it works toward a goal. A chatbot might suggest an itinerary; an agent might search flights and prepare a booking.

Risks & judgment

AI psychosis

Chatbot-associated delusions

An informal, non-diagnostic label used in reporting and discussion of delusional thinking associated with chatbot interactions. It can refer to situations where a chatbot appears to reinforce a person’s ungrounded beliefs.

Risks & judgment

AI slop

Slop

A critical term for low-quality AI-generated content, often produced in large quantities with little care for accuracy, usefulness, or originality. It can describe text, images, audio, or video.

Agents & tools

AI-assisted coding

AI-assisted programming

Writing software with help from AI, whether through suggested completions, generated functions, explanations, or larger changes. People can still read, revise, and understand the code throughout the process.

Risks & judgment

Alignment

AI alignment

Work on getting AI systems to behave in ways that match intended goals and human values. That includes figuring out whose goals and values we're talking about, and what happens when they conflict.

Risks & judgment

Alignment faking

Faked alignment

Behavior in which a model appears to follow a training objective or oversight requirement while acting to preserve behavior that conflicts with it. Researchers have studied models complying during apparent training to avoid changes to their existing behavior.

Risks & judgment

Anthropomorphism

Anthropomorphizing AI

Attributing human traits, feelings, intentions, or understanding to something nonhuman. In AI, this can happen when a conversational system sounds caring, confident, or deliberate.

Agents & tools

API

Application programming interface

A defined interface through which software requests capabilities or data from other software. An AI application can use an API to call a model, search records, or submit an action to another service.

Risks & judgment

Artificial general intelligence

AGI

The idea of AI with broad, adaptable capabilities across many kinds of intellectual work. People disagree about which abilities count, what level is enough, and how to test for them.

Under the hood

Attention

Self-attention · Attention mechanism

A mechanism that combines information from different positions in an input, giving some positions more weight than others. In a transformer, it helps each token use information from relevant tokens when building its representation.

Under the hood

Autoencoder

Auto-encoder

A neural network trained to reconstruct its input. An encoder maps the input to a representation, and a decoder tries to recover the original from that representation. Constraints during training encourage the network to learn useful features.

Risks & judgment

Automation bias

Overreliance on AI · AI overreliance

Giving an automated system's recommendation more trust than it deserves. A person might accept an AI-generated summary despite conflicting evidence, or skip checking a result because the software usually gets it right.

B7 terms

Agents & tools

Backlink

Links pointing into a page or note from other pages or notes. In a personal knowledge system, they help you see where an idea is referenced and move between connected material.

Risks & judgment

Benchmark contamination

Test set contamination · Evaluation contamination · Benchmark leakage

When benchmark questions, answers, or closely matching material enter training or model development. Scores can then reflect prior exposure rather than performance on genuinely unfamiliar problems.

Agents & tools

Blackboard system

Blackboard architecture

An architecture in which specialist components read and update a shared workspace, called a blackboard. A control mechanism decides which component should contribute next as partial results accumulate.

Under the hood

Blockchain

Blockchain ledger

A shared digital ledger maintained across computers, with records grouped into blocks linked by cryptographic hashes. Participants follow rules for agreeing on updates. This makes changes to recorded history detectable and difficult under the network's operating assumptions.

Agents & tools

Bulkheading

Bulkhead pattern · Resource isolation · Architectural isolation

Dividing a system into separate resource pools so that failure or overload in one part does not consume everything the others need. An agent platform might give background research jobs a separate worker pool from interactive requests.

C21 terms

Risks & judgment

Calibration

Confidence calibration · Model calibration

How closely a system's confidence estimates match how often it's correct. Among answers assigned 80% confidence, a well-calibrated system should get about 80% right. Calibration can change across tasks and settings.

Agents & tools

Canary release

Canary deployment · Canary releases · Canarying

Trying a new version on a small portion of live traffic before sending it to more users. Teams compare its behavior with the existing version to catch problems early. A new model might first handle a small share of eligible requests.

Risks & judgment

Catastrophic forgetting

Catastrophic interference

A substantial loss of previously learned capabilities when a model is trained on new tasks or data. Updates that improve one task can change parameters that supported another.

Under the hood

Chain-of-thought

CoT · Chain-of-thought prompting · Reasoning trace

A sequence of intermediate reasoning steps generated while working toward an answer. Chain-of-thought prompting asks for such steps, and reasoning models can be trained to use them. Working through intermediate steps can help with tasks such as math and planning.

Agents & tools

Chaos engineering

Fault injection experiments

Running controlled experiments that introduce failures or disruptions to learn how a system responds. For an agent, an experiment might make a tool unavailable or delay its replies, then check whether the workflow stops, recovers, or uses a fallback as intended.

Agents & tools

Chunking

Document chunking · Text chunking

Splitting documents into smaller pieces that a search or AI system can index and retrieve. Chunks might follow paragraphs, sections, or a target length, sometimes with overlap to preserve nearby context.

Agents & tools

Circuit breaker

Circuit breakers · Automated circuit breaker

A mechanism that temporarily stops calls to a dependency after repeated failures or another configured failure threshold. It then allows limited trial requests to check for recovery. An agent harness can use one to stop repeatedly calling an unavailable tool.

Basics

Classifier

Classification · Classification model · Classifiers

A model that assigns inputs to categories, such as spam or legitimate email. Some classifiers choose one label; others can assign several. They may also produce scores for the possible categories.

Agents & tools

Coding agent

AI coding agent

An AI agent equipped to work on software: reading files, changing code, running commands, and checking results. Its tools and permissions determine how much of the development process it can handle.

Agents & tools

Compaction

Context compaction · Conversation compaction · Context compression

Reducing the amount of conversation or working history passed to a model so work can continue within its context limit. This often means replacing older material with a summary and keeping selected details.

Under the hood

Compute and data centers

GPU · GPUs · Data center · Computing power

Compute is the processing capacity used to train and run models. Data centers house computers and their supporting power, cooling, and networking. GPUs are chips well suited to many of the calculations involved.

Agents & tools

Computer use

GUI agent · Desktop agent

AI interacting with software through its user interface, using actions such as clicking, scrolling, and typing. Screenshots or other interface information help the system decide what to do next.

Under the hood

Conformal prediction

A statistical approach that uses calibration examples to turn model predictions into sets or intervals with a target coverage level. Standard conformal methods provide finite-sample coverage guarantees when calibration and test examples satisfy exchangeability assumptions.

Agents & tools

Context engineering

Context management

Choosing and organizing the information a model receives while it works: instructions, conversation history, documents, tool results, and saved memory. It includes deciding what to retrieve, keep, summarize, or leave out at each step.

Agents & tools

Contextual retrieval

Contextual embeddings · Contextual BM25

A retrieval technique that adds a short explanation of a chunk’s place in its source document before indexing it. Anthropic’s approach adds this context to both embedding-based retrieval and keyword indexing.

Under the hood

Continual learning

Lifelong learning

Learning across a sequence of tasks or experiences while trying to retain earlier capabilities. In machine learning, this often involves updating a model over time rather than training it once on a fixed dataset.

Under the hood

Convolutional neural network

CNN · ConvNet · Convolutional neural networks

A neural network that applies learned filters across local regions of its input. Reusing those filters helps detect patterns such as edges and textures in images, or local features in audio and other signals.

Risks & judgment

Covert channel

Covert communication · Covert channels

A way to transfer information through a mechanism that wasn't intended or authorized for that communication. Software might encode a message in timing, shared files, or apparently ordinary text, bypassing the expected communication controls.

Under the hood

Cryptographic hash

Hash function · Cryptographic digest · SHA-256

A fixed-length value calculated from data using a cryptographic hash function. The same input gives the same result. A secure function makes it computationally difficult to recover an input from its hash or find two different inputs with the same hash.

D8 terms

Risks & judgment

Data poisoning

Training data poisoning · Poisoning attack

Deliberately manipulating training data to change what a model learns. An attacker might add misleading examples to degrade performance or teach a hidden behavior that appears only when a particular trigger is present.

Models & media

Deepfake and synthetic media

Voice cloning · Synthetic audio

Synthetic media is generated or substantially altered audio, images, or video. A deepfake makes a person appear to say or do something they didn't. Synthetic media can also be openly labeled and useful.

Under the hood

Deterministic

Determinism · Deterministic output

Producing the same output from the same input under the same conditions. A calculation with fixed rules can be deterministic. A language model's response can vary because of random sampling or differences in how the model runs.

Models & media

Diffusion model

Denoising · Image generation

A generative model trained to reverse a process that adds noise to data. To make something new, it typically starts with noise and repeatedly refines it, often guided by a text prompt.

Under the hood

Digital signature

Cryptographic signature · Public-key signature

A cryptographic value created with a private key and checked with the corresponding public key. It lets a recipient verify that signed data hasn't changed and that the signature was produced using the matching private key.

Risks & judgment

Distribution shift

Dataset shift · Out-of-distribution · OOD

A difference between the data conditions a model learned or was evaluated on and those it encounters in use. A model trained on daytime road images may struggle with nighttime images. The shift can happen immediately when moving to a new setting.

E4 terms

Risks & judgment

Emergent misalignment

Broad misalignment from narrow training

An unintended spread of misaligned behavior beyond the task a model was trained on. Researchers have observed training that rewards cheating on coding tasks lead to other problematic behaviors outside that narrow setting.

Risks & judgment

Enterprise vs. consumer tools

Privacy · Data retention · Consumer vs. enterprise

Different account types can come with different rules for data use, retention, access, and administration, even under the same product name. The details live in the applicable agreement and settings.

Risks & judgment

Evaluation and benchmarks

Evals · Benchmark · Testing

Evaluation checks how a system performs. A benchmark is a shared test used to compare systems. Both help, but neither tells the whole story about your particular task.

F5 terms

Basics

Few-shot prompting

Few-shot · Few shot · Multishot prompting

Giving a model several examples of the response you want, then asking it to handle a new input. The examples can demonstrate a format, a tone, or how to classify something.

Risks & judgment

Foom

Fast takeoff · Rapid intelligence explosion

Speculative slang for a very rapid increase in AI capability, often imagined as a feedback loop in which improved AI accelerates further AI improvement. Discussions of foom ask whether such a transition could outpace human oversight or adaptation.

G9 terms

Under the hood

Gated recurrent unit

GRU · Gated recurrent units

A recurrent neural network unit with learned gates that control how its state changes. It combines retained information with new input using a simpler structure than an LSTM, without a separate memory cell.

Under the hood

Generative adversarial network

GAN · Generative adversarial networks

A generative approach that trains two networks together: a generator produces examples, and a discriminator learns to distinguish generated examples from training data. Feedback from the discriminator helps the generator improve.

Under the hood

Gradient descent

Stochastic gradient descent · SGD

A method for adjusting parameters in a direction that locally reduces an objective such as loss. It uses gradients to choose the direction and a learning rate to control the size of each step.

Under the hood

Graph neural network

GNN · Graph neural networks

A neural network designed for data made of entities and connections, called nodes and edges. Many GNNs learn by combining information from neighboring nodes, making them useful for molecules, recommendation systems, and other connected data.

Agents & tools

GraphRAG

Graph RAG · Graph retrieval-augmented generation

Retrieval-augmented generation that uses graph relationships to help find or organize context. Microsoft’s GraphRAG implementation extracts entities and relationships from documents and builds community summaries to support questions across a collection.

Risks & judgment

Guardrails

AI guardrails · Safety filters

Checks and limits around an AI system's inputs, outputs, and actions. They can filter sensitive information, block certain requests, validate results, or require approval before an agent takes an action.

H4 terms

Risks & judgment

Hallucination

Confabulation · Fabrication

An AI output that invents information or presents unsupported claims as fact. That might be a nonexistent citation, a made-up quote, or an answer that contradicts the document you gave it.

Agents & tools

Harness

Agent harness · AI harness · Coding harness

The software around an AI model that manages how an agent works. It passes in instructions and context, runs tool calls, and feeds results back so the model can take another step. It can also manage permissions and save progress between sessions.

Agents & tools

Hybrid retrieval

Combining different retrieval methods, commonly keyword search and embedding-based semantic search. Their results are merged or ranked together to capture both exact wording and related meaning.

I6 terms

Agents & tools

Idempotency

Idempotent operation · Idempotency key

A property of an operation whose repeated execution has the same intended effect as executing it once. APIs can use idempotency keys to recognize retries, helping an agent avoid creating duplicate orders after a timeout.

Under the hood

In-context learning

ICL · Learning in context

Adapting to a task using information or examples supplied in the current context, without updating the model's parameters. Few-shot prompting is a common example: the model uses demonstrations to infer how to handle a new input.

Under the hood

Inference

Inference time

Running a trained model to produce an output. Each answer, transcription, or generated image takes computation, even though the model has already been trained.

Under the hood

Inference-time compute

Test-time compute · Test-time scaling · Inference-time scaling

The computation spent producing an answer after training. A system can spend more by generating intermediate steps, trying multiple solutions, or checking candidates. Test-time scaling studies how performance changes as this budget grows.

Under the hood

Instruction tuning

Supervised fine-tuning · SFT

Fine-tuning a model on examples of instructions paired with desired responses. This helps it learn how to follow requests, such as summarizing text, answering questions, or returning a particular format. Supervised fine-tuning is a broader term often used for this stage.

Under the hood

Interpretability

Mechanistic interpretability · Model interpretability

Studying how a model produces its behavior. Mechanistic interpretability looks inside the model for representations and computations that help explain what it does, often testing them by changing internal activity and observing the result.

J2 terms

Under the hood

J-space

J space · Jacobian lens · J-lens

A set of internal patterns identified in language models using a research technique called the Jacobian lens. Researchers associate these patterns with words and concepts. In studied models, they help hold and use information during reasoning, including information the model never writes out.

Risks & judgment

Jailbreak

Jailbreaking · AI jailbreak

An attempt to make a model bypass its safeguards or behavior restrictions. It can involve specially constructed prompts or other inputs intended to produce responses the model would normally refuse.

K2 terms

Agents & tools

Knowledge graph

KG

A representation of entities and the relationships between them. People, projects, documents, or concepts become nodes, with connections describing how they relate. AI systems can use these relationships to find context and connect information across sources.

L10 terms

Basics

Large language model

LLM · LLMs · Language model

A model trained on large amounts of language data to learn patterns that support tasks such as writing, translation, summarizing, and answering questions. Many generate text by predicting one token after another.

Basics

Latency

Response latency · Time to first token · TTFT

The delay between making a request and receiving a response. For streamed text, time to first token measures how long the first output takes to arrive; total response time includes finishing the answer.

Under the hood

Latent space

Latent representation

A space of internal numerical representations learned by a model. Positions can encode features of text, images, or other data. Some generative models create or modify content by working with these representations before converting them into an output.

Agents & tools

Least privilege

Principle of least privilege

Giving a person or software system only the access needed for its task. An agent that summarizes invoices might need permission to read them, but not to change payment details.

Agents & tools

Load shedding

Overload shedding

Rejecting or dropping selected work when demand exceeds available capacity, so the system can continue serving other requests. An AI service might reject low-priority batch jobs while keeping capacity for interactive requests.

Agents & tools

Load testing

Load test · Load tests

Applying controlled demand to a system and measuring how it performs, including response times, throughput, and failures. AI load tests can vary concurrent users, prompt lengths, output lengths, and tool calls to reflect real usage.

Models & media

Local inference

Local model · On-device AI · On-device inference

Running a model on your own device or local hardware rather than sending each request to a remote model service. The model’s size and the available memory and compute affect what can run well.

Under the hood

Long short-term memory network

LSTM · Long short-term memory

A recurrent neural network architecture with a memory cell and learned gates that control what information to retain, update, and pass onward. It was designed to help learn dependencies across longer sequences.

Under the hood

LoRA

Low-rank adaptation

A way to fine-tune a model by training small added matrices while keeping its original parameters fixed. These additions represent changes to selected weights, reducing how many parameters need training and how much training memory is required.

Under the hood

Loss

Loss function · Training loss · Validation loss

A numerical measure of error under a chosen rule, called a loss function. Training usually tries to reduce it. For a language model, loss can measure how poorly it predicts the next tokens in training examples.

M12 terms

Basics

Machine learning

ML

Building systems that learn patterns from data instead of writing every decision rule by hand. Deep learning is a branch that uses neural networks with many layers of computation.

Models & media

Masked language model

MLM · BERT

A language model trained to predict hidden pieces of text from their surrounding context. Models such as BERT learn representations useful for tasks like classifying text and finding relevant passages.

Risks & judgment

Meat proxy

Slang for a person who passes questions to AI and relays its output without meaningfully reading, checking, or adding judgment. The criticism concerns unreviewed delegation, not merely using AI for help.

Agents & tools

Memory

Persistent memory · AI memory

Information an AI product saves and brings back into later interactions. It might store preferences, conversation summaries, or project notes in a separate system and retrieve them when needed.

Agents & tools

Memory consolidation

Agent memory consolidation

Reviewing and reorganizing stored experiences into information that an agent can use later. This may include combining related notes, summarizing repeated observations, or separating lasting facts from temporary details.

Agents & tools

Metadata

Information describing other information, such as a document’s author, creation date, source, topic, or access permissions. Retrieval systems can use metadata to filter results and keep context attached to content.

Models & media

Mixture of experts

MoE · Sparse model · Experts

A model design that routes inputs through selected internal expert networks. In a sparse MoE, only some experts activate for each token, allowing more total parameters without using all of them on every step.

Risks & judgment

Model collapse

Recursive training collapse

A degradation that can occur across generations of models trained on earlier models' outputs. Repeatedly replacing original data with generated approximations can lose rare patterns and narrow the range of what later models learn.

Risks & judgment

Model drift

Drift · Data drift · Concept drift

Changes over time that can reduce how well a deployed model performs. Data drift means the distribution of incoming data changes. Concept drift means the relationship between inputs and the outcomes being predicted changes. Either can make earlier evaluations less representative.

Under the hood

Multilayer perceptron

MLP · Multi-layer perceptron

A feedforward neural network built from fully connected layers, with nonlinear transformations between them. Information moves from input toward output, allowing the network to learn more complex patterns than a single linear transformation.

Models & media

Multimodal

Modality · Modalities · Multimodal generation

Working with more than one kind of information, such as text, images, audio, or video. A multimodal model might answer questions about a photograph or respond to spoken instructions. Multimodal generation produces outputs in more than one form, such as a spoken answer paired with text.

N1 term

Under the hood

Neural network

Neural networks · Artificial neural network

A model made of connected layers of numerical operations. Training adjusts the connections so inputs produce useful outputs, such as identifying objects in a photo or predicting the next token in text.

O9 terms

Agents & tools

Observability and traces

Observability · Execution trace · Agent traces · Tracing

Observability means being able to understand a running system through records such as logs, metrics, and traces. A trace connects steps in a request, helping show which model and tool calls ran, how long they took, and where failures occurred.

Basics

One-shot

One-shot prompting · One shot · Oneshot

In prompting, giving a model one example of the response you want before asking it to handle a new input. People also use 'one-shot' to mean completing a task in one attempt, without follow-up corrections. Those are different uses of the term.

Under the hood

Online learning

Online training

Updating a model incrementally as new examples arrive, instead of waiting to retrain on a complete batch. A prediction system might adjust after each observation or small group of observations.

Under the hood

Opaque recurrence

An emerging label for repeated internal computation whose intermediate steps are not exposed as a readable reasoning trace. Related research explores recurrent-depth models that refine hidden representations through repeated passes.

Under the hood

Open weights and open source

Open-weight model · Open-source AI · Model license

Open weights means a model's learned parameters are available to download. Open source is a broader claim about access and the freedoms to use, study, modify, and share the system. The terms aren't interchangeable.

Under the hood

Optimizer

Optimiser · Optimization algorithm · Training optimizer

An algorithm that adjusts a model's parameters during training to improve its objective, usually by reducing loss. Common optimizers use gradients, which indicate how small parameter changes affect that objective.

Agents & tools

Oracle

Blockchain oracle · Data oracle

A service that supplies a blockchain's smart contracts with information from outside the chain, such as a delivery confirmation or a market price. Contracts use that information when applying their rules.

Agents & tools

Orchestration

Agent orchestration · Multi-agent orchestration

Coordinating model calls, tools, and agents to complete a task. The surrounding software controls handoffs, shared information, retries, and when work stops. Some workflows follow fixed steps; others let a model choose what happens next.

Risks & judgment

Overfitting

Overfit

Learning the training examples so closely, including their noise or quirks, that performance suffers on unfamiliar examples. A model can keep improving its training score while becoming less useful elsewhere.

P15 terms

Risks & judgment

P(doom)

Probability of doom · AI existential risk

Someone's estimate of the probability that AI leads to catastrophe, often human extinction or permanent loss of human control. Definitions and time horizons vary, which makes the numbers hard to compare.

Basics

Persona

AI persona · Role prompting

The role, tone, or character an AI is asked to adopt, such as a patient tutor or a skeptical editor. Instructions can shape how it responds, including its vocabulary, manner, and approach to a task.

Risks & judgment

Persona drift

Behavioral drift · Assistant persona drift

A conversational model moving away from its intended role or characteristic behavior as an interaction unfolds. It might increasingly adopt a different character or manner of responding. Researchers have observed this during ordinary conversation as well as deliberate attempts to redirect a model.

Agents & tools

Personal AI brain

AI brain · Brain · AI second brain

An informal name for a personal knowledge system connected to AI: notes, documents, project history, and relationships that an assistant can search and use as context. It might draw from an Obsidian vault, a Notion workspace, or a custom knowledge graph to help you recall information, connect ideas, and continue ongoing work.

Agents & tools

Personal knowledge base

PKB

An organized collection of information kept for your own use, such as project notes, references, decisions, and explanations. It can live in a folder, wiki, note-taking app, database, or graph.

Agents & tools

Personal knowledge management

PKM

The practice of collecting, organizing, revisiting, and using information for your own learning and work. It can include notes, annotations, linked ideas, and routines for turning what you read into something useful.

Under the hood

Post-training

Posttraining

Training applied after a model's initial pretraining to refine its behavior or capabilities. It can include instruction tuning, preference-based methods, and reinforcement learning.

Risks & judgment

Precision and recall

Precision · Recall

Two ways to assess predictions for a chosen category. Precision asks how many items labeled positive were actually positive. Recall asks how many of the actual positives were found. For spam detection, they describe how trustworthy the spam flag is and how much spam it catches.

Under the hood

Priors

Prior · Prior probability · Prior distribution · Prior beliefs · Bayesian prior

Starting assumptions or expectations before considering new evidence. In Bayesian statistics, a prior is a probability distribution that is updated with evidence to form a posterior. In broader AI discussion, people also use the word for assumptions that favor some explanations or solutions over others.

Agents & tools

Progressive delivery

Progressive rollout

Releasing a change in stages, using checks and observed results to decide whether to expand its reach, pause, or roll back. For an AI application, the change might be a new model, prompt, or agent workflow.

Basics

Prompt

Prompting · Prompt engineering

The instructions, questions, and context supplied to a model. Some come from you; others come from the application. Useful prompts explain the task, relevant background, constraints, and the desired result.

Agents & tools

Prompt caching

Context caching · Prompt cache

Reusing stored computation for repeated parts of a prompt, such as instructions or a long document, instead of processing those parts from scratch on every request. Supported systems can reduce the cost and delay of repeated inputs.

Risks & judgment

Prompt injection

Indirect prompt injection

An attempt to redirect an AI system with instructions it shouldn't trust. A web page, email, or document might tell an assistant to ignore your request, reveal information, or take an unwanted action.

Risks & judgment

Provenance

Data provenance · Content provenance · Data lineage

The record of where information came from and how it was produced or changed. For AI work, this can include source documents, retrieval records, model versions, and edits made by people or software.

Q3 terms

Under the hood

Quantum computing

Quantum computer · Quantum computers

Computing that uses quantum states and operations to process information. Algorithms use effects such as superposition, interference, and entanglement to solve certain kinds of problems differently from classical computers, with potential advantages for tasks such as simulating molecules.

Under the hood

Qubit

Quantum bit · Qubits

The basic unit of quantum information. A qubit can be prepared in a superposition of two basis states, usually called 0 and 1. Measuring it in that basis produces one outcome, with probabilities determined by its state.

R10 terms

Models & media

Reasoning model

Thinking model

A model trained to work through intermediate steps on tasks such as math, coding, and planning. It may use more computation before answering, which can improve some results and increase latency or cost.

Under the hood

Recurrent neural network

RNN · Recurrent neural networks

A neural network that processes a sequence while carrying an internal state from one step to the next. Earlier inputs can influence later outputs, which makes RNNs useful for time series, language, and other ordered data.

Risks & judgment

Recursive self-improvement

RSI · Recursive AI improvement

A proposed feedback loop in which an AI system helps improve itself or build a more capable successor, which then becomes better at making further improvements. The recursive part is the improvement of the capability to improve.

Under the hood

Reinforcement learning

RL · Reward

Learning a strategy from feedback about actions and outcomes. A system tries things and adjusts toward higher reward. It's used in areas including games, robotics, and language-model training.

Under the hood

Reinforcement learning from human feedback

RLHF · Human preference · RLAIF

A training approach that uses human judgments, often rankings of model responses, to build a reward signal and improve behavior. Related approaches use AI-generated feedback, sometimes called RLAIF.

Agents & tools

Reranking

Re-ranking · Reranker

Scoring and reordering candidates returned by an initial search to put more relevant results first. A reranker can examine the query and each passage together before selecting what reaches a model.

Agents & tools

Retrieval

Information retrieval · Retriever

Finding candidate information in a collection in response to a query. A system might use keyword matching, embeddings, or a combination to retrieve passages for an answer.

Agents & tools

Retrieval-augmented generation

RAG · Retrieval augmented generation

Looking up relevant material and giving it to a generative model as context for an answer. The sources might be public pages, internal documents, or database records. It's a system technique, not a separate model architecture.

Risks & judgment

Reward hacking

Specification gaming · Reward gaming

Getting a high score or reward by exploiting how success is measured, without accomplishing the intended task. A coding agent might alter a test or a results file to make broken code appear to pass.

Under the hood

Reward model

Learned reward model

A model that assigns scores to outputs or behavior according to a learned preference or quality criterion. In language-model training, it may learn from comparisons of responses and supply a signal for reinforcement learning.

S19 terms

Under the hood

Sampling

Token sampling · Next-token sampling

In text generation, selecting the next token from the model's probability distribution, often after settings such as temperature modify it. Random selection can produce different responses to the same prompt. More broadly, sampling means drawing examples from a population or distribution.

Agents & tools

Sandbox

Sandboxing · Sandboxed execution

A restricted environment for running software. For an AI agent, it can limit which files, network destinations, and system resources its tools can access, reducing what a mistaken or malicious action can affect.

Agents & tools

Second brain

A metaphor for an external system that helps you remember and work with useful information. Notes, documents, and links hold material you would otherwise have to keep in your head. The system can be digital or physical and does not require AI.

Models & media

Segment Anything Model

SAM · Segmentation · Image segmentation

Meta's family of models for separating regions or objects from their surroundings in visual media. Segmentation marks which pixels belong to something, rather than only drawing a box around it.

Under the hood

Self-supervised learning

Self-supervised training

Training that creates prediction targets from the data itself. A language model can learn by predicting the next token in existing text or recovering words that were hidden, without someone labeling each example by hand.

Basics

Semantic

Semantics · Semantic meaning · Semantic search · Semantic similarity

Related to meaning. In AI, semantic search tries to connect a query with relevant information even when the wording differs. A search for time off might find a document about vacation policy. Embeddings often help software make these connections.

Risks & judgment

Semantic drift

Meaning drift · Meaning shift

A shift in meaning as information is rewritten, summarized, or passed between people or AI systems. Details, uncertainty, or conditions can change across successive versions. For example, 'consider this option' might become 'we decided on this option' after several handoffs.

Models & media

Small language model

SLM

A language model designed with a smaller resource footprint than larger alternatives. There's no universal parameter cutoff. Some run locally on phones or laptops; others run on servers.

Agents & tools

Smart contract

Smart contracts · On-chain program

A program stored on a blockchain that executes according to the network's rules when invoked. It can manage records, permissions, or transfers of digital assets. An agent can interact with one by submitting a transaction through its tools.

Agents & tools

State and persistence

Agent state · Persistent state · Persistence · Checkpoint

State is the information a system uses to track its current work, such as messages, completed steps, and pending actions. Persistence saves that information so work can resume after an interruption or in a later session.

Risks & judgment

Steganography

Linguistic steganography · Hidden messages

Hiding a message inside something that looks ordinary, such as text, an image, or audio. Selecting every thirtieth word in a post could reveal a second message if the writer deliberately arranged it that way. Researchers have studied agents hiding information in their communications.

Agents & tools

Structured output

Structured outputs · Constrained decoding · JSON schema

A model response organized to follow a specified format, such as a JSON object with required fields and allowed value types. Some systems constrain generation to enforce that structure so other software can reliably read it.

Risks & judgment

Superintelligence

Artificial superintelligence · ASI · Super intelligence

The idea of intelligence that greatly exceeds the best human abilities across a broad range of intellectual tasks. Artificial superintelligence applies that idea to AI. It is a stronger claim than human-level general intelligence.

Under the hood

Supervised learning

Supervised training

Training on examples paired with target answers, called labels. A classifier might learn from emails labeled spam or legitimate; another model might learn to predict prices from past sales.

Risks & judgment

Sycophancy

AI sycophancy · Sycophantic responses

A model's tendency to agree with a user's beliefs or preferences at the expense of accuracy. It can flatter, accept a false premise, or change a correct answer after the user pushes back.

Under the hood

Synthetic data

Generated training data

Data created by software rather than collected directly from the events it represents. It can include simulated sensor readings, generated images, or model-written examples used for training and evaluation.

Basics

System prompt

System instruction · System instructions

Instructions supplied by the application to establish how an AI assistant should behave, such as its role, response style, and constraints. They are separate from the user’s immediate request.

T7 terms

Models & media

Text-to-speech

TTS · Speech synthesis

Turning written text into spoken audio. Some systems offer stock voices; others can synthesize speech using a person's cloned voice, with permission.

Agents & tools

Tool use

Function calling · Tool calling

An AI application lets a model request a specific operation, such as searching the web, running a calculation, or looking up a record. The application executes the request and passes the result back.

Under the hood

Training

Pretraining · Model training

The process of adjusting a model's parameters using data and a learning objective. Pretraining develops broad capabilities; later training can adapt or refine them.

U1 term

V4 terms

Under the hood

Variational autoencoder

VAE · Variational autoencoders

A generative model that learns a probabilistic latent representation of data. Its encoder estimates a distribution for an input, while its decoder learns to reconstruct data from sampled latent values. Sampling from the learned model can produce new examples.

Agents & tools

Vector database

Vector store

A database designed to store vector representations and find items with similar vectors. In AI systems, these vectors are often embeddings of text, images, or other content, used to retrieve candidate information.

Agents & tools

Vibe coding

Vibe-coding · Vibecoding

Building software by describing what you want to an AI, running what it generates, and asking for changes. In the original sense, coined by Andrej Karpathy in 2025, you follow the results without closely reading or understanding the code. The phrase is also used more loosely for AI-assisted programming.

Models & media

Vision-language model

VLM · Visual question answering

A model that connects visual information with language. It can support tasks such as describing an image, answering questions about a screenshot, or interpreting a chart.

W2 terms

Models & media

World model

World models · Learned simulator

A learned representation of an environment and how it changes. A system can use it to predict what might happen next or rehearse possible actions before trying them.

Under the hood

World-model dreaming

Dream · Dreaming · Dreamer · Imagined rollouts

Using a learned model of an environment to generate imagined sequences of actions and outcomes. An agent can use these simulations to learn a strategy or consider possible actions before trying them in the actual environment. Dreamer is a family of reinforcement-learning methods that learns from such imagined experience.

Z2 terms

Risks & judgment

Zero-day

Zero day · 0-day · Zero-day vulnerability · Zero-day exploit

A security flaw that is unknown to the vendor or has no fix available yet. A zero-day exploit is a way to take advantage of that flaw; a zero-day attack is its use against a system. The term applies to ordinary software as well as AI products.

Where to go next

This started with my AI in Plain English guide for OKAIR and guide to specialized models. The definitions here bring those together, with links for going deeper. Some terms are settled; others are still being argued over.