AI Glossary

Computer systems designed to perform tasks that typically require human intelligence. This includes learning from experience, recognizing patterns, understanding language, and making decisions.

A type of AI where computers learn from data without being explicitly programmed. The system improves its performance as it processes more examples.

A computer system inspired by how the brain works. It consists of interconnected nodes (artificial neurons) that process information and learn patterns from data.

A subset of machine learning that uses neural networks with multiple layers. It's particularly good at processing complex data like images, audio, and text.

An AI system trained on huge amounts of text data to understand and generate human language. It predicts the next word based on previous words.

The collection of examples and information used to teach an AI system. The quality and diversity of training data significantly affect AI performance.

A step-by-step procedure or set of rules that a computer follows to solve a problem or complete a task.

AI technology that helps computers understand, interpret, and generate human language in a meaningful way.

AI technology that enables computers to interpret and understand visual information from images and videos.

A trained AI system that has learned patterns from data and can make predictions or generate outputs. Think of it as the 'brain' of an AI application.

The adjustable settings or weights in an AI system that determine how it processes information. More parameters generally mean greater capacity to learn complex patterns.

AI systems that can create new content like text, images, music, or code based on patterns learned from training data.

AI systems that classify or categorize existing data rather than creating new content. They learn to distinguish between different types of inputs.

A machine learning approach where AI is trained on labeled examples (input-output pairs) so it learns the correct answers.

A machine learning approach where AI finds patterns in unlabeled data without being told what the correct answers are.

A machine learning approach where an AI learns by interacting with an environment and receiving rewards or punishments for its actions.

The input or question you give to an AI system. A well-written prompt helps the AI understand what you want and provides better results.

The basic unit of text that an AI language model processes. Tokens are typically words or parts of words.

The process of taking a pre-trained AI model and training it further on specific data to make it better at a particular task.

A set of rules that allows different software applications to communicate with each other. Many AI tools offer APIs for developers to use.

When an AI generates false or made-up information that sounds plausible but isn't true. It's a limitation where AI confidently states incorrect facts.

When an AI system shows prejudice or unfair treatment based on patterns in its training data. This can lead to discriminatory outcomes.

When an AI system memorizes specific training examples too well and performs poorly on new, unseen data. It's like memorizing answers instead of learning concepts.

A measure of how often an AI system makes correct predictions or decisions. Higher accuracy means better performance.

A measure of how many of the AI's positive predictions are actually correct. Important when false positives are costly.

A measure of how many actual positive cases the AI correctly identifies. Important when false negatives are costly.

An AI program designed to simulate conversation with users. It can answer questions, provide information, or help with tasks.

An AI system that suggests products, content, or services based on user preferences and behavior patterns.

The ability of AI to determine the emotional tone or opinion expressed in text (positive, negative, or neutral).

The practice of developing and using AI systems responsibly, considering fairness, transparency, accountability, and social impact.