AI (Artificial Intelligence) is technology that enables computers to perform tasks requiring human intelligence, such as recognition, reasoning, and decision-making. AI is designed to analyse data, identify patterns, and can improve over time as it receives more data, often without direct human intervention. AI is used in voice assistants, recommendation systems, self-driving cars, and smart search engines.
Algorithms
Algorithms are step-by-step instructions that a computer follows to solve a problem or perform a task. They are applied in many technologies and applications. AI uses complex algorithms, while simpler algorithms perform tasks like sorting emails or adding numbers without learning or adapting. Note: AI uses algorithms, but not all algorithms are AI.
Generative AI
Generative AI is a type of artificial intelligence capable of actively creating new content—such as text, images, music, or videos—based on the data it was trained on. It is increasingly integrated into the software and systems we use daily.
LLMs (Large Language Models)
A Large Language Model (LLM) is a type of generative AI trained on vast amounts of text to understand, generate, and predict human language. It uses advanced statistical techniques (like neural networks and transformer architectures) to process textual input and provide meaningful, contextual responses. LLMs can answer questions, summarize texts, conduct conversations, translate, write code, or generate creative writing (such as poems or stories). Note: LLMs are a type of generative AI, but generative AI is broader than just LLMs.
Prompt
A prompt is a textual instruction or question you give to an AI system to generate a specific answer, result, or creation. A prompt guides the behavior of the AI model. The clearer and more specific the prompt, the better (and more targeted) the result. You can specify style, tone, or format (e.g., "in the style of Shakespeare," "in bullet points," "as a haiku"). Prompt tuning is the deliberate experimentation to better steer an AI model like an LLM by systematically improving or optimizing the prompt, so the model produces increasingly better or more specific output for a given task.