Glossary
A
AI Hallucination
AI hallucinations occur when ChatGPT generates information that is incorrect or nonsensical, but it presents it as if it were accurate. This can happen due to limitations in the training data or model.
Algorithm
An algorithm is a set of rules or instructions that an AI system follows to solve a problem or perform a task. In AI, algorithms help in processing data and making decisions based on patterns.
API (Application Programming Interface)
An API is a set of protocols that allow different software systems to communicate with each other. The ChatGPT API enables developers to integrate the AI into various applications, websites, or systems to provide conversational AI services.
Artificial General Intelligence (AGI)
AGI refers to a theoretical form of AI that can perform any intellectual task a human can, across any domain. Unlike narrow AI, which is designed for specific tasks, AGI would have the ability to generalize knowledge across different areas.
Artificial Intelligence
Artificial intelligence refers to a computer or a robot controlled by a computer that can carry out tasks typically linked to human intelligence. It involves the science and engineering of creating smart machines, particularly intelligent software systems. AI enables computers, robots, or software to mimic intelligent human thinking and behaviour.
Artificial Neural Network (ANN)
An artificial neural network is a computational model inspired by the human brain's structure, consisting of layers of interconnected nodes. It is used in various AI tasks, including image processing and speech recognition.
Autocomplete
Autocomplete is a feature in many AI models where the system predicts and completes sentences or phrases based on the input provided. ChatGPT uses this concept to predict your next prompt and generate relevant responses.
Avatar
An avatar is a digital representation of a user or AI. In ChatGPT, you might use an avatar to personalise your interaction or see a visual representation of who you’re speaking with in some applications.
B
Bias
In AI, bias refers to the tendency of models to produce skewed or imbalanced results due to flaws in the training data. ChatGPT strives to minimise bias, but its responses can sometimes reflect unintended patterns present in the data it was trained on.
Big Data
Big data refers to extremely large data sets that are difficult to process using traditional methods. AI systems often rely on big data to train models, enabling them to identify patterns, make predictions, and generate insights.
C
Chat History
Chat history refers to the record of your past interactions with ChatGPT. It allows the model to remember previous exchanges during a conversation, helping maintain context and coherence in ongoing dialogues.
Chatbot
A chatbot is an AI system that interacts with people through text or voice. ChatGPT is a type of chatbot that responds to your prompts, answering questions and helping with tasks.
ChatGPT
ChatGPT is a state-of-the-art language model developed by OpenAI designed to generate human-like text based on the prompts it receives. It excels in a variety of tasks, including answering questions, writing essays, creating fictional stories, and even generating images using integrated tools like DALL-E. The model is trained on a vast array of internet text, allowing it to produce responses that are often insightful, relevant, and contextually appropriate.
Cognitive Computing
Cognitive computing refers to AI systems that simulate human thought processes. These systems aim to understand natural language, recognise patterns, and learn from data to provide insights and support decision-making.
Computer Vision
Computer vision is a branch of AI that enables machines to interpret and understand visual information from the world, such as images and videos. It’s used in applications like facial recognition and self-driving cars.
Confidence Score
A confidence score indicates how certain ChatGPT is about the response it generates. While not always visible, higher confidence typically correlates with more accurate answers, while lower confidence might indicate uncertainty.
Context (Important)
In conversations with ChatGPT, context refers to the ongoing understanding of the previous interactions. Maintaining context allows the AI to give more coherent and relevant responses as the conversation progresses, improving the flow of dialogue.
D
DALL-E
DALL-E is an AI model developed by OpenAI that generates unique images from text descriptions. It uses deep learning techniques to understand and create visual content based on detailed prompts, enabling creative applications such as designing artwork, visual storytelling, and concept generation from language inputs.
Data Mining
Data mining is the process of discovering patterns, trends, and useful information from large data sets. AI uses data mining techniques to find relevant insights and improve its decision-making capabilities.
Deep Learning
Deep learning is a subset of machine learning that uses neural networks with many layers to model complex patterns in large amounts of data. It powers many AI systems, including image recognition and natural language processing.
Default Settings
Default settings are the pre-configured options that come with ChatGPT or any software. They help you get started, but you can change them to suit your preferences or needs.
E
Edge AI
Edge AI refers to AI processing that takes place locally on a device rather than in the cloud. This allows for faster decision-making and reduced data transfer, making it ideal for real-time applications like autonomous vehicles and smart cameras.
Ethical AI
Ethical AI refers to the responsible development and deployment of AI systems to ensure they operate fairly, transparently, and without harm. ChatGPT follows ethical guidelines designed to prevent harmful or biased responses, safeguarding users during interactions.
Expert System
An expert system is an AI program that uses a database of knowledge and a set of rules to mimic the decision-making abilities of a human expert in a specific domain, such as medical diagnosis or financial analysis.
F
Fallback Response
A fallback response is a general or neutral reply that ChatGPT gives when it is unsure of how to respond to a particular prompt. It helps ensure the conversation continues even when the AI cannot generate a precise answer.
Fine-tuning
Fine-tuning is the process of adjusting an AI model to improve its performance in specific tasks or domains. ChatGPT can be fine-tuned on specialised data to enhance its ability to generate responses tailored to particular industries or topics.
Fuzzy Logic
Fuzzy logic is a form of logic used in AI systems that deals with reasoning that is approximate rather than fixed and exact. It allows machines to handle uncertain or imprecise information, simulating more human-like decision-making.
G
Generative AI
Generative AI is a subset of artificial intelligence focused on creating new content, such as text, images, or music, based on input data. ChatGPT is an example of generative AI, where it generates text responses based on prompts provided by the user.
GPT (Generative Pre-trained Transformer)
GPT is an advanced AI model designed to understand and generate human-like text. It learns from vast amounts of data and uses deep learning techniques to generate coherent, contextually relevant responses to prompts, making it ideal for tasks like writing, conversation, and problem-solving.
H
Human-in-the-loop (HITL)
Human-in-the-loop refers to the practice of involving humans in the process of refining and guiding AI outputs. In some cases, human reviewers might provide feedback or corrections to improve the responses generated by ChatGPT.
I
Input
Input refers to the information you give to ChatGPT, like a question or a command. This input helps guide the AI in generating an appropriate response. It’s more than just a Prompt.
Interaction
An interaction refers to the back-and-forth exchange between you and ChatGPT. Each interaction includes a prompt and response, forming a conversation that evolves depending on your input and the AI’s ability to understand and reply.
Iterations
Iterations refer to the cycles of improvement made to AI models through updates and feedback. As ChatGPT evolves, new iterations are released to enhance its functionality, accuracy, and conversational abilities.
J
Journey
In AI and ChatGPT usage, a journey refers to the sequence of interactions or steps a user takes from their first prompt to the final response. It can also refer to learning or exploring new features over time.
K
Knowledge Representation
Knowledge representation is a field of AI that focuses on how to represent information about the world in a form that a computer can use to solve complex tasks, such as diagnosing diseases or understanding language.
L
LLM
An LLM (Large Language Model) is a type of AI designed to process and generate natural language by learning from vast datasets. It uses deep learning techniques to understand context, predict text, and generate coherent responses, enabling applications like chatbots, content creation, and language translation.
M
Machine Learning (ML)
Machine learning is a type of AI that enables systems to learn and improve from experience without being explicitly programmed. ChatGPT uses ML algorithms to predict and generate responses based on patterns it learned from large amounts of text data.
Model Version
The model version refers to the specific iteration of the AI you're interacting with, such as GPT-3.5 or GPT-4. Each version offers improvements in performance, accuracy, and the complexity of responses based on advancements in the model's training.
Multimodal AI
Multimodal AI refers to models that can process and generate more than one type of data, such as text, images, or audio. Although ChatGPT primarily handles text, future versions may incorporate multimodal capabilities to enhance interactions.
N
Natural Language Processing (NLP)
NLP is the branch of AI that enables machines to understand, interpret, and generate human language. It powers ChatGPT by allowing it to comprehend and produce text that mimics human conversation, making interactions more natural and intuitive.
Natural Language Understanding (NLU)
NLU is a component of NLP (Natural Language Processing) that focuses on a machine’s ability to comprehend and interpret human language in a meaningful way. It helps AI understand context, intent, and meaning in conversations.
Neural Network
Neural networks are computational models inspired by the human brain, consisting of layers of interconnected nodes (neurons). They are the backbone of AI systems like ChatGPT, enabling it to process vast amounts of data and generate intelligent responses.
Knowledge Cutoff
The knowledge cutoff is the date beyond which an AI model like ChatGPT has not been updated with new information. For ChatGPT, this date is usually indicated, and it means the AI might not know events, developments, or facts that occurred after that time.
O
Output
Output is the information or response that ChatGPT gives back after processing your input. It's the answer, explanation, or solution that the AI generates based on what you asked.
Overfitting
Overfitting occurs when an AI model learns the details and noise of the training data too well, causing it to perform poorly on new, unseen data. This issue can limit ChatGPT's ability to generalise and respond accurately to a wide range of prompts.
P
Persona (Important)
A persona in ChatGPT refers to the character or role that the AI adopts during a conversation. By defining a persona, you can guide ChatGPT to respond with a specific tone, style, or expertise, making interactions more personalised and aligned with your needs or preferences.
Prediction
Prediction in AI is the process of guessing what comes next based on past data. In ChatGPT, it predicts the next word or sentence in a conversation to give a relevant response.
Predictive Analytics
Predictive analytics is the use of AI and machine learning to analyse historical data and make predictions about future events. It is widely used in industries such as finance, marketing, and healthcare for forecasting and decision-making.
Privacy Settings
Privacy settings allow you to control the data shared during your interactions with ChatGPT. Understanding these settings helps you ensure that your personal information and conversations are protected.
Prompt (Important)
A prompt is the input or question you provide to ChatGPT to initiate a conversation. It can range from simple queries to more complex instructions, guiding the AI on what type of response you're expecting. The quality of your prompt often determines the relevance of the AI’s response.
Q
Query
A query is another word for a question or request that you give to ChatGPT. It's the prompt that starts the conversation and guides the AI in providing an answer or information.
R
Real-time Processing
Real-time processing is the ability of AI systems like ChatGPT to generate responses almost instantaneously as prompts are received. This quick turnaround makes the interaction feel like a natural conversation.
Reinforcement Learning
Reinforcement learning is a method where an AI learns to make decisions by taking actions in an environment and receiving feedback in the form of rewards or penalties. It is used in scenarios like game playing and robotics.
Response (Important)
A response is the output generated by ChatGPT based on the given prompt. It is the AI’s reply, which can include information, explanations, suggestions, or solutions. The clarity and detail in your prompt influence the accuracy and usefulness of the response.
Robotics
Robotics is a field that intersects with AI, involving the design, construction, and operation of robots. AI enables robots to perform tasks autonomously or semi-autonomously by interpreting data and making decisions.
Roleplay
Roleplay in ChatGPT refers to its ability to take on personas or simulate conversations based on specific scenarios. You can ask ChatGPT to act as a teacher, assistant, or expert in a certain field to receive specialised responses.
S
Safety Filters
Safety filters are built-in mechanisms designed to prevent ChatGPT from generating harmful, offensive, or inappropriate content. These filters monitor interactions and help ensure a safe user experience.
Session
A session refers to the time you spend interacting with ChatGPT from start to finish. Each session can involve multiple prompts and responses, and keeping track of session context helps the AI generate relevant answers.
Singularity
The singularity is a hypothetical point in the future where AI surpasses human intelligence and begins to improve itself at an exponential rate, potentially leading to transformative changes in society.
Speech Recognition
Speech recognition is the ability of AI to understand spoken language and turn it into text. Some versions of ChatGPT can take voice input and respond in text.
Supervised Learning
Supervised learning is a type of machine learning where an AI model is trained on labeled data. The model learns to map inputs to the correct outputs by being shown examples of input-output pairs.
System Message
A system message refers to the invisible instructions guiding ChatGPT's behaviour. These messages help the AI understand the context of the conversation, ensuring that responses align with the expected tone, format, and purpose.
T
Temperature Setting
Temperature refers to a parameter that controls how random or creative the AI's responses are. A higher temperature value leads to more creative and diverse outputs, while a lower value results in more focused, predictable answers.
Text-to-Speech
Text-to-speech is a feature that allows ChatGPT to read its responses aloud. This can make interactions easier, especially for users who prefer listening over reading.
Toggle
A toggle is a switch or button that turns a feature on or off. In ChatGPT settings, you might find toggles to adjust features like text length or response creativity.
Token (Important)
A token is a unit of text that AI processes. It can represent a word, a part of a word, or punctuation. ChatGPT handles text by breaking it into tokens to generate responses. The more tokens used in a prompt or response, the more detailed or complex the exchange becomes.
Training Data
Training data refers to the large sets of information used to teach AI models like ChatGPT. This data helps the AI learn patterns and language structures and how to generate accurate responses. The more diverse and extensive the training data, the better the AI's performance.
Turing Test
The Turing Test is a measure of a machine’s ability to exhibit intelligent behaviour equivalent to, or indistinguishable from, that of a human. If an AI passes the Turing Test, it means it can hold a conversation without revealing it’s not human.
U
Unsupervised Learning
In unsupervised learning, an AI model is given data without explicit labels and is tasked with finding patterns or relationships within the data. This approach is often used for clustering or dimensionality reduction tasks.
User Interface
The user interface is the visual layout and design that you interact with when using ChatGPT. It includes features like text boxes, buttons, and menus, making it easier for you to input prompts and read responses.
V
Voice Input
Voice input allows you to speak to ChatGPT instead of typing. The AI recognises your voice and converts it into text to generate a response, making interaction more convenient for some users.
W
Wizard
A wizard in software refers to a simple, step-by-step guide that helps you complete a task. In some versions of AI tools, a wizard might guide you through setting up your account or understanding basic features.
X
XML (Extensible Markup Language)
While not specific to ChatGPT, XML is a language used to structure data for sharing between systems. In the context of AI, it can be used to format inputs or outputs for various applications.
Y
YouTube Integration
Some AI platforms, including those similar to ChatGPT, can integrate with YouTube, allowing you to interact with videos or get AI-generated summaries and insights based on video content.
Z
Zoom Integration
In certain setups, ChatGPT can be integrated with platforms like Zoom. This allows for AI assistance during virtual meetings, such as generating summaries, transcriptions, or assisting in real-time conversations.