# AI Glossary

URL: https://www.onsomble.ai/tools/ai-glossary

Plain-language definitions for the AI terms that actually matter, with practical context on why each one is relevant.

## Categories

- Models & Platforms — Large language models, foundation models, and the platforms that serve them.
- Retrieval & Data — Techniques for grounding AI in external knowledge and structured data.
- Agents & Orchestration — Autonomous AI agents, multi-step workflows, and orchestration frameworks.
- Evaluation & Safety — Testing, red-teaming, guardrails, and responsible AI practices.
- Developer Experience — Tools, SDKs, and patterns that make building with AI easier.
- Interfaces & UX — How AI shows up in products — chat, copilots, generative UI.
- AI Search & Discoverability — Terms covering Answer Engine Optimisation (AEO), AI search behaviour, and how brands show up — or don't — in AI-generated answers.

## AI Agent

An AI system that can autonomously plan, reason, and take actions to accomplish goals, often using tools and external APIs.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/ai-agent)

## AI Alignment

The challenge of ensuring AI systems act in accordance with human intentions and values — making them do what we actually want, not just what we literally ask for.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/ai-alignment)

## AI Bias

Systematic unfairness in AI outputs caused by skewed training data, flawed labelling, or model design choices that reflect and amplify existing societal inequities.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/ai-bias)

## AI Citation

An AI citation is a single instance of an AI assistant mentioning a brand by name (or linking to its content) in a generated answer — the unit of measurement that AEO platforms aggregate into visibility scores.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/ai-citation)

## AI Orchestration

The coordination and management of multiple AI models, tools, data flows, and human inputs within a unified workflow to accomplish complex tasks that no single component could handle alone.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/ai-orchestration)

## AI Overviews

AI Overviews are Google's AI-generated answer summaries that appear above the traditional list of search results, launched broadly in May 2024 and rolled out across most query types since.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/ai-overviews)

## AI SEO

AI SEO is an industry-popular term for the practice of optimising content to be cited by AI assistants — essentially a synonym for AEO and GEO, but framed as an extension of traditional SEO rather than a separate discipline.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/ai-seo)

## AI Visibility

AI visibility is the measurable presence and accuracy of a brand inside AI-assistant responses — covering how often it's mentioned, in what tone, with what facts, and against which competitors.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/ai-visibility)

## Agentic AI

AI systems that can autonomously plan, reason, use tools, and execute multi-step tasks with minimal human oversight — going beyond simple question-answering to take actions on behalf of users.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/agentic-ai)

## Analytical AI

AI systems that find patterns, anomalies, and insights in large datasets, used for tasks like fraud detection, medical imaging analysis, and business intelligence.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/analytical-ai)

## Answer Engine

An answer engine is an AI system that responds to user questions with synthesised answers rather than a list of links — examples include ChatGPT, Perplexity, Gemini, Google AI Overviews, and Microsoft Copilot.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/answer-engine)

## Answer Engine Optimisation (AEO)

Answer Engine Optimisation (AEO) is the practice of making a brand discoverable, accurately described, and frequently cited by AI assistants like ChatGPT, Gemini, Perplexity, and Google's AI Overviews.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/answer-engine-optimisation)

## Artificial Intelligence (AI)

A broad field of computer science focused on building systems that can perform tasks normally requiring human intelligence, such as understanding language, recognizing patterns, and making decisions.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/artificial-intelligence)

## Chain-of-Thought Reasoning (CoT)

A prompting technique that improves LLM accuracy on complex tasks by guiding the model to show its reasoning step-by-step before arriving at a final answer.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/chain-of-thought-reasoning)

## ChatGPT

ChatGPT is OpenAI's conversational AI assistant — the consumer product that ignited the generative AI category in November 2022 and reached 900 million weekly active users by February 2026.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/chatgpt)

## Chunking

The process of splitting documents into smaller, semantically meaningful segments optimised for embedding and retrieval in AI systems like RAG pipelines.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/chunking)

## Citation Authority

Citation authority is the ranked list of third-party domains that AI assistants actually quote when answering questions about a topic — the AI-search equivalent of a backlink profile.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/citation-authority)

## Claude

Claude is Anthropic's family of large language models and the chat product that exposes them — built around safety-first design, longer context windows, and Constitutional AI alignment.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/claude)

## Compound AI Systems

AI architectures that combine multiple models, retrievers, tools, and control logic to tackle tasks that no single model could reliably handle on its own.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/compound-ai-systems)

## Context Window

The maximum amount of text (measured in tokens) that a language model can consider at once — including both the input prompt and the generated output.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/context-window)

## Conversational AI

AI systems designed to engage in natural language dialogue with humans, ranging from simple chatbots with scripted responses to advanced assistants powered by large language models.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/conversational-ai)

## Copilot

An AI assistant embedded directly into a workflow tool (IDE, browser, email) that suggests actions, generates content, or automates tasks inline.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/copilot)

## Crawlability

Crawlability is whether — and how easily — automated agents (search crawlers, AI bots) can fetch and read the content of a website, governed by robots.txt, response codes, JavaScript rendering, paywalls, login walls, and rate limits.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/crawlability)

## Data Privacy in AI

The set of concerns and practices around protecting personal and sensitive information when using AI systems, covering training data, user inputs, model outputs, and data retention.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/data-privacy-in-ai)

## Deep Learning (DL)

A subset of machine learning that uses neural networks with many layers to learn increasingly abstract representations of data, powering breakthroughs in language, vision, and generation.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/deep-learning)

## Diffusion Model

A generative AI architecture that creates images, video, and other media by learning to gradually remove noise from random static until a coherent output emerges.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/diffusion-model)

## Embeddings

Dense numerical representations of text (or other data) in a high-dimensional vector space, where similar meanings are placed closer together.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/embeddings)

## Few-Shot Prompting

A prompting technique where you include a few examples of the desired input-output format in your prompt, helping the model understand exactly what you want without any fine-tuning.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/few-shot-prompting)

## Fine-tuning

The process of further training a pre-trained model on a smaller, task-specific dataset to improve its performance on that particular task or domain.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/fine-tuning)

## Function Calling

The mechanism that allows LLMs to interact with external tools and APIs by outputting structured data — typically JSON — specifying which function to invoke and with what parameters.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/function-calling)

## GPTBot and AI Crawlers

GPTBot is OpenAI's web crawler — one of a growing list of AI-specific crawlers (PerplexityBot, ClaudeBot, Google-Extended, Applebot-Extended, etc.) that fetch content for training and live retrieval by AI assistants.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/gptbot)

## Gemini

Gemini is Google's family of large language models and the conversational AI product that exposes them — also the engine behind Google's AI Overviews and AI Mode in Search.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/gemini)

## Generative AI (GenAI)

Artificial intelligence that creates new content — text, images, video, audio, or code — by learning patterns from existing data and producing original outputs in response to prompts.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/generative-ai)

## Generative Engine Optimization (GEO)

Generative Engine Optimization (GEO) is the academic name for AEO, formalised in a 2024 KDD paper by researchers at Princeton and IIT Delhi who tested which content modifications improve visibility in AI-generated answers.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/generative-engine-optimization)

## Google AI Mode

Google AI Mode is the full conversational AI search experience inside Google Search — an evolution of AI Overviews into a multi-turn chat interface that lets users follow up on complex questions.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/google-ai-mode)

## Guardrails

Programmatic constraints placed around AI model inputs and outputs to prevent harmful, off-topic, or policy-violating behavior.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/guardrails)

## Hallucination

When an AI model generates information that sounds plausible but is factually incorrect, fabricated, or unsupported by its training data or provided context.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/hallucination)

## Hybrid Search

A retrieval approach that combines traditional keyword matching (BM25) with semantic vector search to capture both the precision of exact term matches and the contextual understanding of meaning-based search.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/hybrid-search)

## Knowledge Cutoff

The date beyond which a language model has no training data, meaning it cannot know about events, discoveries, or changes that occurred after that point.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/knowledge-cutoff)

## Knowledge Graph (KG)

A structured representation of information as a network of entities and their relationships, enabling machines to reason about connections between concepts.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/knowledge-graph)

## LLMO (LLMO)

LLMO (Large Language Model Optimisation) is one of several names for the practice of getting cited by AI assistants — interchangeable with AEO, GEO, and AI SEO, but specifically frames the target as the underlying language model rather than the answer-engine product.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/llmo)

## Large Language Model (LLM)

A neural network trained on massive text corpora that can understand and generate human language, typically with billions of parameters.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/large-language-model)

## MLOps (MLOps)

The set of practices and tools for deploying, monitoring, and maintaining machine learning models in production — essentially DevOps principles applied to the ML lifecycle.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/mlops)

## Machine Learning (ML)

A subset of AI where systems learn patterns from data rather than following explicitly programmed rules, improving their performance as they see more examples.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/machine-learning)

## Microsoft Copilot

Microsoft Copilot is Microsoft's AI assistant family powered by GPT models and Bing search — embedded across Windows, Microsoft 365, GitHub, Edge, and as a standalone chat product at copilot.microsoft.com.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/microsoft-copilot)

## Mixture of Experts (MoE)

A neural network architecture that scales model capacity efficiently by routing each input through only a small subset of specialized sub-networks ("experts"), keeping compute costs manageable even as total model size grows.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/mixture-of-experts)

## Model Context Protocol (MCP)

An open standard for connecting AI assistants to external data sources and tools through a unified, composable interface.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/model-context-protocol)

## Multimodal AI

AI systems that can process, understand, and generate across multiple types of data — text, images, audio, video, and code — within a single model.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/multimodal-ai)

## Next-Token Prediction

The core mechanism of large language models: given a sequence of text, predict the most likely next piece (token), then repeat to generate coherent text one token at a time.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/next-token-prediction)

## Observability

The practice of monitoring, tracing, and evaluating AI system behavior in production — including LLM calls, latency, costs, retrieval quality, and output correctness.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/observability)

## Perplexity

Perplexity is an AI-native answer engine that always cites its sources — built from the ground up around web search and inline citations, rather than retrofitting a chat product to add browsing.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/perplexity)

## Pre-training

The initial training phase where a language model learns general language patterns from a massive text corpus, before being fine-tuned for specific tasks or behaviors.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/pre-training)

## Predictive AI

AI systems that forecast outcomes based on historical data patterns, used for tasks like demand forecasting, risk assessment, and recommendation engines.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/predictive-ai)

## Prompt Engineering

The practice of designing and refining inputs to language models to elicit more accurate, useful, and consistent outputs.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/prompt-engineering)

## RLHF (RLHF)

A training technique where human preferences are used to fine-tune a language model through reinforcement learning, teaching it to produce responses that humans judge as helpful, accurate, and safe.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/rlhf)

## Reasoning Models

LLMs trained with reinforcement learning to "think before they answer" by generating internal chains of reasoning — producing more accurate results on complex tasks like maths, coding, and multi-step logic at the cost of higher latency and token usage.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/reasoning-models)

## Red Teaming

Systematic adversarial testing of AI systems to identify vulnerabilities, failure modes, and unintended behaviours before deployment — adapted from cybersecurity to probe AI-specific weaknesses like prompt injection and jailbreaks.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/red-teaming)

## Reinforcement Learning (RL)

A machine learning approach where an agent learns by taking actions in an environment and receiving rewards or penalties, gradually discovering which strategies produce the best outcomes.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/reinforcement-learning)

## Reranking

A second-stage retrieval technique that re-scores and reorders an initial set of retrieved documents using a more computationally expensive cross-encoder model to surface the most relevant results.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/reranking)

## Retrieval-Augmented Generation (RAG)

A technique that grounds a language model's output in external data by retrieving relevant documents before generating a response.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/retrieval-augmented-generation)

## Schema Markup

Schema markup is structured metadata embedded in a webpage (using the schema.org vocabulary, typically as JSON-LD) that tells search engines and AI assistants exactly what a page is about — product, article, FAQ, organisation, person, recipe, event, etc.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/schema-markup)

## Semantic Search

A search technique that understands the meaning and intent behind queries rather than matching exact keywords, using vector embeddings to find conceptually relevant results even when different words are used.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/semantic-search)

## Sentiment Analysis

In an AI-search context, sentiment analysis is the classification of how AI assistants describe a brand — positive, neutral, or negative — across the answers they generate, beyond just whether the brand is mentioned.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/sentiment-analysis)

## Share of Voice (SoV)

Share of voice (in the context of AI search) is the percentage of AI-generated answers about a category that mention a given brand, measured against named competitors over the same prompt set.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/share-of-voice)

## Small Language Model (SLM)

A compact AI language model — typically under 10 billion parameters — designed to run efficiently on edge devices and single GPUs while delivering strong task-specific performance.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/small-language-model)

## Structured Data

Structured data is any machine-readable representation of information on a webpage — most commonly schema.org JSON-LD, but also microdata, RDFa, sitemaps, OpenGraph tags, Twitter cards, and (increasingly) llms.txt files.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/structured-data)

## Structured Output

A technique for constraining a language model's output to follow a specific format like JSON, XML, or a defined schema, ensuring the response can be reliably parsed by downstream code.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/structured-output)

## Supervised Learning

A machine learning approach where the model learns from labeled examples — input-output pairs where the correct answer is provided during training.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/supervised-learning)

## System Prompt

The foundational instruction set given to an LLM that defines its role, behaviour, tone, and constraints for a particular application — set once at the application level and shaping all subsequent user interactions.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/system-prompt)

## Temperature

A parameter that controls how random or deterministic an LLM's output is — lower values produce more predictable, focused responses while higher values increase creativity and variation.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/temperature)

## Token

The basic unit of text that a language model processes — typically a word, subword, or punctuation mark, roughly equivalent to 3/4 of an English word.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/token)

## Topic Clustering

Topic clustering is the grouping of related prompts, AI answers, or content pieces into thematic clusters — used in AEO to consolidate prompt-level visibility data into actionable narratives about how AI describes a category.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/topic-clustering)

## Training Data

The dataset used to teach a machine learning model, containing the examples and patterns the model learns to recognize and reproduce.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/training-data)

## Transformer Architecture

A neural network architecture that powers modern AI by processing entire input sequences simultaneously through an attention mechanism, rather than reading them word by word.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/transformer-architecture)

## Unsupervised Learning

A machine learning approach where the model finds patterns and structure in data without labeled examples, discovering groupings and relationships on its own.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/unsupervised-learning)

## Vector Database

A database optimized for storing, indexing, and querying high-dimensional vector embeddings, enabling fast similarity search at scale.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/vector-database)

## Zero-click Search

A zero-click search is a search query that ends without the user clicking through to any external website — the user gets their answer from the search results page itself.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/zero-click-search)

## llms.txt

llms.txt is an emerging standard for a plain-text file at the root of a website that summarises the site for AI assistants — analogous to robots.txt for crawlers or sitemap.xml for search engines.

[Read the full definition](https://www.onsomble.ai/tools/ai-glossary/llms-txt)
