ChatGPT
ChatGPT is an AI assistant developed by OpenAI that generates conversational responses and helps with tasks using large language models.
Large language model: A neural network trained on large text collections to predict and generate sequences of language. ChatGPT uses large language models to produce and interpret conversational text.
Conversational agent: A software system designed to interact with people through dialogue. ChatGPT is a general-purpose conversational agent rather than a single-task chatbot.
GPT-4: A multimodal large language model developed by OpenAI and released in 2023. GPT-4 is one model family used to power versions of ChatGPT.
Generative artificial intelligence: Artificial intelligence systems that create new content, such as text, images, audio, or code. ChatGPT became a widely used example of generative AI for text and dialogue.
Transformer: A neural-network architecture that uses attention to process relationships among sequence elements. Transformer networks form the core architecture of the models behind ChatGPT.
Code generation: The creation of computer-program code from specifications, examples, or natural-language instructions. ChatGPT can draft, explain, and revise code in response to requests.
ChatGPT Enterprise: OpenAI's ChatGPT offering designed for organizations, with enterprise-oriented controls and features. It adapts the ChatGPT service for organizational use and administration.
AI literacy: The knowledge and skills needed to understand, evaluate, and use artificial intelligence systems. Using ChatGPT well requires judging its capabilities, uncertainty, and limitations.
Instruction tuning: Training a model on examples of instructions paired with desired responses. It helps turn a text predictor into a system that follows user requests.
Machine translation: The automated conversion of text or speech from one language into another. ChatGPT can translate language as one use of its text-generation abilities.