A-Z of artificial intelligence

Agentic AI

Artificial intelligence that can independently plan and execute tasks to achieve goals with minimal human intervention.

AGI (Artificial General Intelligence)

A hypothetical type of AI that can understand, learn, and apply intelligence to solve any problem that a human can, across a wide range of tasks and domains.

AI (Artificial Intelligence)

The simulation of human intelligence processes by machines, especially computer systems. These processes include learning, reasoning, and self-correction.

Algorithm

A set of well-defined instructions or a step-by-step procedure for solving a problem or performing a computation.

ANN (Artificial Neural Network)

A computational model inspired by the structure and function of biological neural networks, used for machine learning.

Anomaly Detection

The identification of rare items, events, or observations which raise suspicions by differing significantly from the majority of the data.

API (Application Programming Interface)

A set of defined rules that enable different software applications to communicate with each other.

Bias

In machine learning, a systematic error in a model's prediction due to incorrect assumptions in the learning algorithm or data. Can lead to unfair or inaccurate results.

Big Data

Extremely large datasets that may be analysed computationally to reveal patterns, trends, and associations, especially relating to human behaviour and interactions.

Black box

A term used to describe an AI system whose internal workings are not easily understandable or interpretable by humans.

Chatbot

A computer program designed to simulate human conversation with users, either through text or voice interactions.

Clustering

An unsupervised machine learning task of grouping a set of objects in such a way that objects in the same group (called a cluster) are more similar to each other than to those in other groups.

CNN (Convolutional Neural Network)

A specialised type of neural network particularly effective for image and video analysis.

Coding

The act of creating a set of instructions in a programming language that a computer can understand and follow in order to execute specific tasks.

Cognitive Computing

A subfield of AI that aims to simulate human thought processes, including reasoning, learning, and self-correction, in complex systems.

Computer Vision

A field of AI that enables computers to "see" and interpret visual information from the real world, such as images and videos.

Cyberattack

A deliberate attempt by an individual or group to gain unauthorised access to, damage, disrupt, or steal data from computer systems, networks, or digital devices. Examples include phishing, malware, data poisoning and Distributed Denial-of-Service (DDoS) attacks.

Data Mining

The process of discovering patterns, insights, and knowledge from large datasets using various analytical techniques.

Data poisoning

A type of cyberattack in which an attacker intentionally corrupts or manipulates the data used to train a machine learning or AI system, causing it to make incorrect or harmful decisions.

Deep Learning

A subfield of machine learning that uses artificial neural networks with multiple layers (deep neural networks) to learn complex patterns from large amounts of data.

Deepfake

Synthesised media in which a person's face or voice is digitally altered to resemble someone else, often using deep learning techniques.

De-skilling

The reduction in human skills or expertise because tasks are increasingly performed or assisted by artificial intelligence systems.

DNN (Deep Neural Network)

An artificial neural network with multiple hidden layers, which enables it to learn more complex representations of data.

ELIZA Effect

The tendency of people to unconsciously assume computer responses are more intelligent or human-like than they actually are, even when knowing they are interacting with a program.    

Expert Systems

Computer programs designed to mimic the decision-making ability of a human expert in a specific field.    

Explainable AI (XAI)

A field of AI research focused on making AI models more transparent and understandable to humans, addressing the "black box" problem.    

Fuzzy Logic

A form of many-valued logic in which the truth values of variables may be any real number between 0 and 1, inclusive, used to deal with approximate reasoning.

GAN (Generative Adversarial Network)

A class of AI algorithms that consists of two neural networks, a generator and a discriminator, which compete against each other to generate new data that resembles the training data.

General AI (Strong AI)

AGI (Artificial General Intelligence): A hypothetical type of AI that can understand, learn, and apply intelligence to solve any problem that a human can, across a wide range of tasks and domains.

Generalisation

The ability of a machine learning model to perform well on unseen data, beyond the specific examples it was trained on.

Generative AI

A type of artificial intelligence that can create original content, such as text, images, music, or code, based on patterns it has learned from existing data.

GPT (Generative Pre-Trained Transformer)

A type of AI model that understands and generates human-like text by learning patterns from large datasets. It can be used to answer questions and summarise information, create text, generate code and translate languages. GPT models can power tools like chatbots, helping them respond in a way that feels conversational.

Hallucination

In generative AI models, the generation of plausible but factually incorrect or nonsensical information.

Heuristic

A practical, experience-based approach to problem-solving that is not guaranteed to be optimal or perfect but is often sufficient for immediate goals.

Human-in-the-loop

A model where humans actively participate in a system's workflow, training or decision-making process. Humans provide critical thinking, ethical judgement and ultimate approval, thus ensuring meaningful oversight of automated processes.

Hyperparameters

Parameters whose values are set before the learning process begins, controlling aspects of the learning algorithm itself (e.g., learning rate, number of layers).

Inference

The process of using a trained AI model to make predictions or decisions on new, unseen data.    

Intelligent Agent

An autonomous entity that perceives its environment and takes actions that maximise its chances of achieving its goals.    

IoT (Internet of Things)

A network of physical objects embedded with sensors, software and other technologies for the purpose of connecting and exchanging data with other devices and systems over the internet.    

Knowledge Representation

The study of how to represent information about the world in a form that a computer system can use to solve complex tasks.

Language Model

A statistical model that determines the probability of a sequence of words, used in various NLP tasks like text generation and translation.    
Large Language Model

A type of language model with a very large number of parameters, trained on massive amounts of text data, enabling it to perform various language understanding and generation tasks.    

Learning Rate

A hyperparameter in machine learning that controls how much the model's weights are adjusted with respect to the loss gradient during training.    

Linear Regression

A basic supervised learning algorithm that models the relationship between a dependent variable and one or more independent variables by fitting a linear equation to the observed data.    

Machine Learning (ML)

A subset of AI that enables systems to learn from data without being explicitly programmed.

Machine Translation (MT)

A subfield of computational linguistics that translates text or speech from one natural language to another using AI.

Malware

Short for malicious software and refers to any program or file designed to gain unauthorised access to, harm, exploit, or disrupt computers, networks, or devices.

Model

In machine learning, a mathematical representation of a real-world process or data, trained to make predictions or decisions.

Narrow AI (Weak AI)

Artificial intelligence that is designed to perform a specific task or a limited range of tasks but cannot operate beyond its programmed purpose.

Natural Language Generation (NLG)

A subfield of NLP focused on generating human-like text from structured data or other inputs.

Natural Language Processing (NLP)

A field of AI that focuses on enabling computers to understand, interpret, and generate human language.

Pattern Recognition

The automated recognition of patterns and regularities in data, often used in conjunction with machine learning.

Phishing

A type of cyberattack where attackers use fake emails, messages or websites to trick people into giving away personal or sensitive information, such as passwords or bank details, by pretending to be a trustworthy source.

Planning and optimisation

The ability of an AI system to find the best sequence of actions or decisions to achieve a specific goal efficiently.

Prompt Engineering

The art and science of crafting effective prompts or instructions to guide generative AI models to produce desired outputs.

Recurrent Neural Network (RNN)

A type of neural network where connections between nodes form a directed graph along a sequence, allowing it to process sequential data like text or time series.

Regression

A machine learning task of predicting a continuous numerical output.

Reinforcement Learning (RL)

A type of machine learning where an agent learns to make decisions by interacting with an environment, receiving rewards for desirable actions and penalties for undesirable ones.

Robot

A machine capable of carrying out a complex series of actions automatically, especially one programmable by computer.

Robotics

The interdisciplinary field concerned with the design, construction, operation, and use of robots.

Scraping

An automated process of extracting large amounts of data from websites.    

Sentiment Analysis

The process of computationally identifying and categorising opinions expressed in a piece of text, especially in order to determine the writer's attitude toward a particular topic or product.    

Singularity (Technological Singularity)

A hypothetical future point in time when technological growth becomes uncontrollable and irreversible, resulting in unfathomable changes to human civilisation.    

Strong AI

See AGI (Artificial General Intelligence).    

Superintelligent AI

A hypothetical form of artificial intelligence that surpasses human intelligence in all areas, including reasoning, problem-solving, creativity, and decision-making.    

Supervised Learning

A type of machine learning where the model is trained on labelled data, meaning the input data is paired with the correct output.    

Symbolic AI

An approach to AI that focuses on representing knowledge and reasoning using symbols and logical rules, rather than statistical methods.    

Tensor

A multi-dimensional array of numbers. In deep learning, data is often represented as tensors.

TensorFlow

An open-source machine learning framework developed by Google, widely used for building and training neural networks.

Training

The process of teaching an artificial intelligence system how to perform tasks by feeding it data and allowing it to learn patterns from that data.

Training Data

The dataset used to train a machine learning model.

Transfer Learning

A machine learning technique where a model trained on one task is repurposed or fine-tuned for a second, related task.

Turing Test

A test of a machine's ability to exhibit intelligent behaviour equivalent to, or indistinguishable from, that of a human.

Unsupervised Learning

A type of machine learning where the model learns from unlabelled data, identifying patterns and structures without explicit guidance.    

Validation Set

A portion of the data used to evaluate the performance of a machine learning model during training and to tune hyperparameters.

Weak AI

See ANI (Artificial Narrow Intelligence).

Weight

In neural networks, a parameter that determines the strength of the connection between two neurons. Weights are adjusted during the training process.