AI Glossary – Key Terms & Acronyms

Term	Definition
AI	Artificial Intelligence – Computer systems that simulate human intelligence.
LLM	Large Language Model – An AI model trained on vast text data to understand and generate human-like language.
ML	Machine Learning – A subset of AI focused on systems that learn from data to improve performance over time.
NLP	Natural Language Processing – The ability of AI to interpret, understand, and generate human language.
DL	Deep Learning – A subset of ML using neural networks with many layers for complex pattern recognition.
GPT	Generative Pre-trained Transformer – A type of LLM architecture developed by OpenAI, used in models like ChatGPT.
API	Application Programming Interface – A way for software systems to communicate; AI models are often accessed via APIs.
Prompt Engineering	The practice of crafting effective inputs (prompts) to guide AI responses or outputs.
Training Data	The datasets used to teach AI/ML models how to perform specific tasks.
Fine-Tuning	Additional training of an AI model on specific data to specialize it for a particular use case.
Token	A chunk of text (word or word part) that AI processes. Models like GPT-4 have token limits (e.g., 8k or 32k tokens).
Inference	The process of using a trained model to generate an output (e.g., answer, prediction, or text).
Bias (AI Bias)	Systematic errors in AI outputs caused by skewed training data or flawed algorithms.
Hallucination	When an AI model confidently generates factually incorrect or fictional information.
Chatbot	An AI system designed to simulate conversation with human users.
Vector Embeddings	Mathematical representations of words, sentences, or data that preserve meaning/context in AI systems.
RAG	Retrieval-Augmented Generation – An AI method that combines generation (LLM) with real-time retrieval of external data.
Zero-shot Learning	The ability of a model to solve tasks without specific training examples, relying only on general knowledge.
Few-shot Learning	A model’s ability to learn and adapt with only a few examples provided in the prompt.
Agent	A system that can perform tasks autonomously, often combining multiple AI capabilities (e.g., planning, action-taking).
TTS	Text-to-Speech – AI that converts written text into spoken audio.
STT / ASR	Speech-to-Text / Automatic Speech Recognition – AI that transcribes spoken language into written text.
Image Generation	AI models that create images based on text prompts (e.g., DALL·E, Midjourney).
Computer Vision	AI systems that interpret and analyze visual data from the world (e.g., images, video).
OpenAI	The research organization behind GPT models, ChatGPT, DALL·E, Codex, and others.
Fine-tuned Model	An AI model that has been specially adapted using targeted training data for niche tasks or industries.
Multi-modal AI	AI that can process and combine different data types (text, image, video, audio, etc.).
Synthetic Data	Artificially generated data used to train or test AI systems when real data is unavailable or sensitive.