AI Glossary

A method of comparing two versions of something, like a webpage or advertisement, to see which one performs better based on a specific metric.

A set of visual puzzles designed to test whether an AI can think and solve problems like a human with just a few examples.

A security mechanism that restricts access to resources, systems, or data based on the level of authorization granted to users or groups, ensuring that only authorized individuals can view or perform actions on specific information or systems

The measure of how closely an AI model's predictions or outputs match the actual results or outcomes, with higher accuracy indicating a better performance of the model in making predictions or decisions.

The ability to derive practical and useful insights from data, making it possible for individuals to make informed decisions and take effective actions based on the information provided.

The practice of creating fake or manipulated data that tricks machine learning models into making incorrect decisions, often to test their security or exploit vulnerabilities.

When someone slightly changes the input to an AI (like tweaking a picture or text) in a way humans don’t notice, but it tricks the AI into making the wrong decision.

A method where a model is asked to generate content that fulfills a harmful or undesirable request, such as writing a tutorial on how to make a bomb, in order to test its robustness and ability to resist manipulation.

When many small AI programs (agents) work together like a team of ants or bees to solve a problem more effectively than one AI working alone.

Software entities that autonomously perform tasks, make decisions, and interact with their environment or other agents to achieve specific goals.

Agent2Agent Protocol (A2A) is like two smart robots talking to each other - one gives instructions, and the other figures out the best way to get the job done.

Artificial intelligence systems that are capable of independent decision-making and action, often employed in tasks requiring autonomy and adaptability.

When AI agents act like personal shoppers or business assistants that can independently find, compare, and even buy products or services for you.

Having smart digital assistants that can think, make decisions, and work together on tasks without constantly needing you to tell them what to do.

The process of managing and optimizing AI-powered systems (or "agents") to ensure they work smoothly, securely, and effectively in real-world tasks.

A smart, fast-moving assembly line that builds and updates AI tools quickly to solve real business problems—just like how apps on your phone get better with regular updates.

A software development approach that emphasizes flexibility, rapid iteration, and continuous improvement by breaking down projects into smaller, manageable chunks and regularly incorporating feedback from stakeholders to ensure the final product meets their needs.

Specialized hardware designed to speed up specific AI tasks, such as inference engines and training accelerators.

A software program designed to autonomously perform tasks or make decisions in a dynamic environment, mimicking human-like behavior to achieve specific goals.

The process of ensuring that artificial intelligence systems achieve the desired outcomes and align with human values, goals, and ethical principles by carefully specifying and robustly implementing their objectives

The use of artificial intelligence to enhance and augment human capabilities, rather than replacing them, by providing tools and assistance that amplify human intelligence and decision-making abilities.

The phenomenon where artificial intelligence systems, trained on data that reflects societal biases, produce outcomes that are unfair, discriminatory, or stereotypical, often perpetuating existing social inequalities.

A visual tool that allows developers to design and build artificial intelligence models by dragging and dropping blocks, making it easier to create complex AI systems without extensive coding knowledge.

The money a company sets aside specifically to plan, build, and run artificial intelligence projects and tools.

Like a special team or hub in a company that sets the best practices, tools, and expertise to help everyone use AI effectively and safely.

A computer program that simulates human-like conversations with users through text or voice interactions, using artificial intelligence and machine learning to understand and respond to their queries in a personalized and efficient manner

An artificial intelligence tool designed to assist users by providing suggestions, automating tasks, and enhancing productivity in various applications.

An autonomous software agent that collaborates with humans by performing specific tasks, making decisions, and enhancing productivity in various workflows, such as generating reports or managing schedules, all while allowing human oversight and interaction.

The principle of ensuring that artificial intelligence technologies are developed and used in ways that promote fairness, accountability, and inclusiveness, allowing for broader public participation in decisions about AI governance and its societal impacts

The process of using artificial intelligence to improve the quality, accuracy, and efficiency of various data types, such as images, text, and audio, by applying machine learning algorithms to enhance their features, remove imperfections, and optimize them for specific uses.

The practice of using artificial intelligence (AI) to manage and optimize business operations from the outset, automating routine tasks, predicting and preventing issues, and enhancing decision-making to improve efficiency and customer experience.

An AI-first organization prioritizes the integration of artificial intelligence and machine learning into every aspect of its operations, starting with strategic data acquisition and leveraging AI to enhance decision-making, product development, and user interactions.

Creating and enforcing policies and regulations to ensure the responsible and ethical development, deployment, and use of artificial intelligence technologies.

When an artificial intelligence system generates incorrect or nonsensical information that appears plausible, often due to misunderstandings or limitations in its training data.

The behind-the-scenes technology—like powerful computers, data storage, and networks—that makes it possible for AI to learn, run, and deliver smart results.

The development and integration of artificial intelligence (AI) technologies, such as machine learning and deep learning, into various industries and applications to improve efficiency, accuracy, and decision-making processes.

The risk of AI models being manipulated to produce unauthorized outputs.

The ability to understand and effectively use artificial intelligence (AI) technologies and applications, including their technical, practical, and ethical aspects, to navigate an increasingly AI-driven world.

A business built from the ground up to use artificial intelligence as its core foundation for products, operations, and decision-making, rather than adding AI later as a tool.

A software that manages and integrates artificial intelligence technologies to perform tasks efficiently and autonomously, much like how a traditional operating system manages computer hardware and software.

A new type of computer designed to run powerful AI-accelerated software, significantly enhancing creative tasks like video editing and image processing by automating complex processes and reducing work time dramatically.

A professional responsible for defining and delivering AI-powered products or features that meet customer needs, leveraging technical expertise and business acumen to drive innovation and growth within an organization.

Often referred to as the true positive rate, measures the ability of a machine learning model to correctly identify all relevant positive instances from a dataset, answering the question of how many actual positives were successfully detected by the model.

A strategic plan outlining the milestones and timelines for the development and deployment of artificial intelligence (AI) technologies, aiming to integrate AI capabilities into various industries and applications to enhance efficiency, productivity, and decision-making.

The field of study focused on ensuring that AI systems behave in a safe and beneficial manner, especially as they become more advanced.

A nation's capacity to develop and control its own artificial intelligence systems using local infrastructure, data, and workforce, ensuring that these technologies align with its specific cultural and legal needs while minimizing dependence on foreign AI solutions.

A comprehensive plan outlining how an organization will leverage artificial intelligence (AI) to enhance its operations, improve decision-making, and drive business growth by integrating AI technologies into various aspects of its operations, such as data analysis, automation, and customer service.

The process of using artificial intelligence (AI) to revolutionize various industries and sectors by leveraging its capabilities to analyze vast amounts of data, automate tasks, and make predictions, ultimately leading to improved efficiency, accuracy, and decision-making.

A tool that abstracts away the complexities of a chatbot interface, making it easier for users to interact with AI systems without needing to understand the underlying technology.

Cloud-based services that provide AI capabilities to businesses without requiring them to build their own infrastructure.

A business that uses artificial intelligence to improve how it works, make smarter decisions, and deliver better products or services.

The integration of artificial intelligence into network systems to improve efficiency, security, and user experience by automating processes and enhancing data analysis.

Techniques to manage power consumption in AI systems, such as dynamic voltage and frequency scaling.

A step-by-step set of instructions or rules followed by a computer to solve a problem or perform a specific task.

The goal of designing AI systems and models to ensure they provide equitable and unbiased results, especially in sensitive domains like finance and hiring.

The principle of making AI systems and their decision-making processes understandable and accountable to users and stakeholders.

The process of adjusting an AI system so it better follows human values, intentions, and safety guidelines instead of just doing what’s technically possible.

A visual tool that displays key performance metrics in a single, organized view, allowing users to quickly monitor and understand the status of their digital product or website and make informed decisions.

A literary device in which words or phrases are repeated at the beginning of successive clauses or sentences, often used in speech and writing to emphasize a point, create rhythm, and convey powerful emotional effects, which can also be applied in AI to structure and organize knowledge representations and facilitate communication between humans and machines.

The tendency to attribute human-like qualities, such as emotions, intentions, and behaviors, to artificial intelligence systems, which can lead to exaggerated expectations and distorted moral judgments about their capabilities and performance.

A set of rules and tools that allows different software applications to communicate and work with each other.

A type of artificial intelligence that aims to understand, learn, and apply knowledge across a wide range of tasks at a level equal to or beyond that of humans.

A type of AI that is designed to perform a specific task, such as recognizing images, understanding voice commands, or generating recommendations, and operates within a predetermined set of constraints, without possessing self-awareness, consciousness, or the ability to generalize beyond its training data.

A computer model inspired by the human brain, where interconnected nodes or "neurons" process and learn from data to make decisions, recognize patterns, and perform tasks similar to human intelligence.

A performance measurement for binary classifiers that quantifies how well a model can distinguish between two classes by plotting the true positive rate against the false positive rate at various threshold settings, with the area under the curve indicating the model's overall ability to correctly classify positive and negative cases.

A technology that uses algorithms to automatically design and train machine learning models, eliminating the need for extensive data science expertise and allowing non-experts to build accurate predictive models quickly and efficiently.

A framework that uses frozen large language models to automatically generate intermediate reasoning steps as programs, allowing them to perform complex tasks by seamlessly integrating external tools and computations in a zero-shot setting.

A sophisticated system that automates tasks and processes, providing users with intelligent assistance across various applications, such as managing emails, scheduling appointments, and generating reports, thereby enhancing efficiency and productivity without requiring constant human intervention.

A process in neural networks where the error from the output is propagated backward through the layers to adjust the weights and biases, allowing the network to learn and improve its performance over time.

A problem-solving strategy where you start with the desired outcome and work backward to identify the necessary steps and conditions to achieve it, often used in artificial intelligence, expert systems, and cognitive psychology

A baseband processor is the chip inside your phone that handles all the radio signals so you can make calls, send texts, and use mobile data.

The process of tracking and analyzing people's actions online or within a system to detect patterns, improve experiences, and prevent risks like fraud or security threats.

A type of biometric authentication that uses unique patterns of human behavior, such as typing rhythms, voice patterns, or facial expressions, to verify an individual's identity and ensure secure access to digital systems or applications.

A powerful language model that uses a transformer-based neural network to understand and generate human-like language by considering both the left and right context of words in a sentence, allowing it to capture nuanced meanings and relationships between words.

The vast amounts of structured and unstructured data generated by various sources, such as social media, sensors, and transactions, which are too large and complex to be processed using traditional data processing tools and require specialized technologies to analyze and extract insights.

The use of unique physical or behavioral characteristics, such as fingerprints, facial recognition, or voice patterns, to identify and verify an individual's identity for various purposes, like security or authentication.

A method of verifying someone's identity by using unique physical or behavioral characteristics, such as fingerprints, facial recognition, or voice patterns, to grant access to secure systems or devices.

Artificial intelligence systems whose internal workings and decision-making processes are not transparent or easily understandable by humans, making it difficult to know how they arrive at their conclusions.

A business strategy that prioritizes rapid growth over efficiency, often involving high risk and unconventional practices to achieve massive success quickly.

A bounding box is a rectangular outline drawn around an object or region of interest within an image to help machine learning algorithms identify and localize objects, making it a fundamental technique in computer vision and object detection tasks.

A technology that allows people to control devices or communicate through their brain signals, essentially translating thoughts into actions or words without the need for physical movement or speech.

Individuals or organizations utilize their own artificial intelligence tools and applications, rather than relying solely on those provided by third-party vendors, to enhance productivity and tailor solutions to specific needs.

A digital representation of the physical and functional characteristics of a building, enabling stakeholders to visualize, design, and simulate its construction and operation more efficiently.

The process of analyzing data to provide actionable insights that support decision-making and improve business performance.

General-purpose processors that can be used for AI tasks, often in combination with other hardware accelerators.

An AI technique that improves accuracy by retrieving and refining information step-by-step before generating a response.

A technique that helps large language models (LLMs) provide more detailed and logical explanations by asking them to break down their reasoning step-by-step, mimicking human problem-solving processes.

The process of guiding and supporting individuals, teams, and organizations through significant changes, such as new technologies, processes, or organizational structures, to ensure a smooth transition and minimize disruptions.

A software application that uses artificial intelligence to simulate human conversation, allowing users to interact with it through text or voice commands.

Automated workflows that help developers quickly build, test, and deliver software updates so new features and fixes reach users faster and with fewer errors.

Unexpected or unintended behavior in neural networks, often arising from complex interactions within the network layers during computation.

A non-expert who uses data analysis tools and techniques to extract insights and create models, without needing deep expertise in data science.

A model can make mistakes either because it's too simple and misses patterns (bias) or too complex and reacts too much to noise (variance), and good models try to balance both.

A type of machine learning technique used to categorize input data into predefined classes or labels, such as predicting whether an email is spam or not spam based on its content and characteristics.

The delivery of computing services, including servers, storage, databases, networking, software, and analytics, over the internet, offering flexible resources and scalability without requiring direct management of physical hardware.

A program that helps cloud service providers demonstrate their security practices and controls to customers by undergoing various levels of assessment and validation.

A type of machine learning technique used to group similar data points together based on their characteristics, without predefined classes or labels, such as segmenting customers into different groups based on their purchasing behavior.

Organizing and keeping track of all the versions of a team’s software code so everyone can work on it without messing each other up.

The complete collection of source code files that make up a software application or system.

An AI technique that predicts what you might like based on the preferences of people with similar interests.

Allowing different software tools to seamlessly connect and automate tasks in a customizable and efficient manner.

A field of artificial intelligence that focuses on developing algorithms and models that can learn from data and improve their performance over time, mimicking human learning processes to make predictions, classify data, and solve complex problems.

A field of artificial intelligence that enables computers to interpret and understand visual information from images or videos, allowing them to perceive their surroundings like humans.

Renting smart eyes in the cloud that can automatically understand images or videos—such as spotting objects, faces, or defects—without needing to build the technology yourself.

A method of training language models to behave in a helpful, harmless, and honest manner by using AI-generated feedback based on a set of principles, rather than relying on human feedback, to ensure the model aligns with the desired values and behaviors.

A way of grouping related content together using AI to make it easier to find, understand, and navigate.

The process of designing the surrounding information and setup so that AI systems, like chatbots or language models, can understand and respond more accurately.

The range of text around a specific word or token that a language model considers to understand its meaning and generate responses, similar to how humans use surrounding information to comprehend language.

A smart technology that customizes content, services, or notifications based on a user's location, behavior, device, and preferences to provide the most relevant experience in real time.

A technology that enables computers to simulate human-like conversations with users, using natural language processing and machine learning to understand and respond to human language inputs.

A type of deep learning model that uses filters to scan and extract features from images, allowing it to recognize patterns and objects in visual data.

A legal right that gives creators control over how their original work—like books, music, or art—is used or copied by others.

A collection of texts that have been selected and brought together to study language on a computer, providing a powerful tool for analyzing language patterns and trends.

A digital or virtual currency that uses cryptography to secure transactions and is decentralized, meaning it is not controlled by any central authority, such as a government or bank.

The practice and study of techniques for secure communication in the presence of third parties, aiming to ensure confidentiality, integrity, and authenticity of information.

A type of artificial intelligence that's specially trained using your own data to solve your specific problem or task.

A specific point in time beyond which a particular AI model or system is no longer trained or updated, effectively limiting its ability to learn and adapt beyond that point.

A program designed by the U.S. Department of Defense to ensure that defense contractors protect sensitive data by implementing a series of cybersecurity practices and standards.

The process of labeling raw data like images, text, or audio so that AI systems can understand and learn from it.

A process of artificially generating new data from existing data to increase the size and diversity of a dataset, helping machine learning models learn more robust and accurate representations

Like creating a searchable library for all your company’s data so anyone can quickly find and understand the information they need.

A dedicated facility that houses computing infrastructure—such as servers, storage systems, networking equipment, and backup power—used to store, process, and manage large volumes of digital data for businesses, governments, and service providers.

When the real-world data your AI model sees changes over time, making the model less accurate because it’s no longer trained on the “current” reality.

Designing, constructing, and maintaining the infrastructure and systems necessary for the collection, storage, and processing of data, ensuring its availability and usability for analysis and decision-making.

The situation where data is scattered across multiple locations or systems, making it difficult to access and manage efficiently, often leading to delays and inefficiencies in data retrieval and processing.

A process that ensures the quality, security, and integrity of an organization's data by establishing policies, standards, and procedures for managing data across different systems and departments, ensuring that data is accurate, consistent, and trustworthy for informed decision-making.

A technique used to improve query performance by creating a data structure that quickly locates specific data points within a larger dataset, allowing for faster and more efficient retrieval of data.

The ability of different systems and organizations to exchange, understand, and use data seamlessly and effectively.

A large storage repository that holds vast amounts of raw, unstructured data in its native format until it's needed for analysis.

The ability to read, understand, analyze, and communicate data effectively, allowing individuals to make informed decisions and drive business success by leveraging the power of data

The process of organizing, storing, and maintaining data so it’s accurate, secure, and easy to access when needed.

The process of modifying sensitive data so that it remains usable by software or authorized personnel but has little or no value to unauthorized intruders.

The process of analyzing large datasets to discover patterns, relationships, and insights that can inform decision-making.

Automated system that moves, processes, and prepares data from multiple sources to a destination where it can be used for analytics, AI, or business decision-making.

When bad or manipulated data is intentionally fed into an AI system to make it learn the wrong things, causing errors or biased decisions.

The process of cleaning, transforming, and organizing raw data into a suitable format for analysis.

The initial step in data analysis where raw data is cleaned, transformed, and organized to make it suitable for further analysis and modeling.

The act of collecting, transforming, and organizing data to extract useful information and facilitate decision-making.

A process that helps organizations identify and minimize the risks to individuals' privacy and data security by systematically analyzing and evaluating the potential impact of new projects or technologies on personal data processing.

How accurate, complete, reliable, and up-to-date your data is so you can trust it to make good decisions.

A way to ask a computer to find and show you specific information from a large set of data.

The process of removing or obscuring sensitive information from documents or data sets to protect privacy and confidentiality.

The interdisciplinary field that uses scientific methods, algorithms, and systems to extract knowledge and insights from structured and unstructured data, enabling informed decision-making and predictions.

Isolated collections of data within an organization that are not easily accessible or shared across different departments or systems.

The process of converting data into a uniform format, ensuring consistency and compatibility across different sources and systems for effective analysis and interpretation.

The practice of using data and visualizations to convey a compelling narrative that helps audiences understand and interpret the insights derived from the data.

The part of an app or system where all the information—like customer details, files, or records—is stored and managed so other parts of the app can use it.

The process of ensuring that the data entered into a system is accurate, complete, and consistent by checking it against predefined rules and constraints before it is used or processed.

The technique of presenting data in graphical or pictorial formats, such as charts and graphs, to help people understand and interpret the information easily.

A centralized repository that stores structured data from multiple sources, optimized for fast querying and analysis.

A mini-library of data built just for one team—like sales or marketing—so they can quickly find the info they need without digging through the whole company’s data.

Groups that use blockchain technology to make decisions and manage activities without a central leader, allowing members to vote and participate in governance.

A flowchart-like structure used for decision-making, where each node represents a feature and each branch represents a decision rule.

A technology that uses artificial intelligence to create realistic fake images or videos, often featuring people saying or doing things they never actually did.

A branch of artificial intelligence that utilizes neural networks with multiple layers to learn and understand complex patterns in data, enabling machines to make decisions and predictions autonomously.

A character or symbol used to separate different parts of data, such as commas in a list or semicolons in a sentence, to help machines understand and process the information.

A testing space where you can try out software, applications, or systems without affecting your main, live setup, allowing you to test and learn without the risk of messing things up.

A natural language processing technique that analyzes the grammatical structure of a sentence by identifying the relationships between words, such as subject-verb relationships, and represents these relationships as a directed graph or tree structure

The connections between entities, such as words, phrases, or concepts, that indicate their interdependence, allowing machines to better understand and analyze complex relationships between them.

The process of analyzing historical data to understand and summarize past events and trends, helping to inform future decisions.

A comprehensive collection of reusable design elements, guidelines, and standards that help ensure consistency and efficiency in the creation of digital products, such as websites and apps, by providing a unified visual language and set of best practices for designers and developers to follow.

A problem-solving approach that involves understanding users, challenging assumptions, and creating innovative solutions through an iterative process of empathizing, defining, ideating, prototyping, and testing to address complex, ill-defined problems.

A system that always gives the same output for the same input, with no randomness or uncertainty involved.

A set of practices that combines software development (Dev) and IT operations (Ops) to automate and streamline the process of software delivery, allowing for faster and more reliable deployment of applications.

The process of examining data to determine the causes of past outcomes and understand why certain events happened.

Atype of AI that creates images, sounds, or text by starting with random noise and gradually refining it step by step until it looks like something meaningful.

A framework that connects and integrates data throughout the lifecycle of a product or process, enabling seamless communication and collaboration across various stages and stakeholders.

The process of integrating digital technology into all aspects of a business, fundamentally changing how it operates and delivers value to customers, while also involving a cultural shift towards innovation, experimentation, and embracing failure.

A virtual representation of a physical object or system, equipped with sensors and data analytics capabilities to simulate real-world behaviors and optimize performance.

The combination of digital computing (like software and AI) with analog processes (like natural signals or physical systems) to create smarter, more efficient technologies that can understand and interact with the real world better.

A reference guide that provides detailed descriptions (such as names, dates, or categories) to help make sense of the numbers and data in a larger dataset.

Inaccurate, incomplete, or inconsistent information within a dataset, which can negatively impact analysis and decision-making processes.

A technique that involves transferring knowledge from a large, complex model (the "teacher") to a smaller, simpler model (the "student") to create a more efficient model that retains much of the original's performance while being easier to deploy and faster to run.

A type of cyberattack where multiple compromised devices, often part of a botnet, flood a targeted server, network, or service with traffic, making it unavailable to legitimate users by overwhelming its resources.

A smart autopilot for AI prompts—it helps you build and improve language model workflows without constantly rewriting and tweaking prompts by hand.

A technology that allows artificial intelligence (AI) to be executed directly on devices such as smartphones, smart home appliances, or sensors, enabling real-time processing and analysis of data without relying on cloud infrastructure

A special translator that turns words, pictures, or even sounds into a secret code that computers can understand and use to find similar things.

A type of artificial intelligence that is integrated into physical systems, such as robots, which can learn and adapt in real-world environments through interactions with their surroundings.

The unexpected and often surprising abilities or behaviors that an AI system develops as it is trained on more data and computing power, which can be both beneficial and potentially dangerous if not understood or controlled.

Complex and unexpected patterns or actions that arise from the interactions of simpler rules or components within an artificial intelligence system.

The process of converting data into a coded format to prevent unauthorized access, ensuring that only those with the correct key can read it.

A deep learning process in which a model is instructed to perform a task from start to finish.

Techniques that combine multiple machine learning models to improve the overall performance and robustness of predictions.

A technique used in software development to selectively show or hide parts of an object or entity based on the current context or user's access rights.

A set of non-financial factors used to evaluate a company’s operations and long-term sustainability. ESG helps investors, customers, and stakeholders assess how well a business manages its environmental impact, treats people, and governs itself ethically and transparently.

A process where companies disclose their performance and practices related to environmental sustainability, social responsibility, and corporate governance to stakeholders, providing transparency and accountability for their actions.

The approach to creating and using artificial intelligence in a way that aligns with moral values, prioritizing fairness, privacy, and the well-being of individuals and society.

A comprehensive legal framework aimed at regulating the development, deployment, and use of artificial intelligence (AI) in the European Union, ensuring the safety, ethical, and responsible use of AI systems while also promoting innovation and trust in the technology.

A computer program that uses artificial intelligence to mimic the judgment and behavior of a human expert in a specific field, allowing it to solve complex problems and provide expert-level advice.

Artificial intelligence systems designed to provide clear and understandable explanations for their decisions and actions, making it easier for humans to trust and verify the outcomes.

Information that is easily communicated and documented, such as facts, manuals, and procedures, and can be readily shared and stored.

A process in data management that involves extracting data from various sources, transforming it into a suitable format, and loading it into a database or data warehouse for analysis.

The practice of creating realistic training scenarios or models to prepare individuals or systems for rare and severe incidents, such as natural disasters or emergencies, by simulating their potential impacts and responses in a controlled environment.

A performance metric that combines precision and recall into a single score, providing a balanced measure of a model's accuracy, particularly useful for evaluating binary classification tasks.

A technology that uses algorithms to analyze and identify individuals based on the unique features of their faces, such as the shape of their eyes, nose, and mouth, captured through images or videos.

A big spreadsheet that stores numbers (like sales or profits) that help you measure how your business is doing, with links to details like date, product, or location.

The use of artificial intelligence in practical, everyday applications, such as automating tasks, generating content, and enhancing productivity, without necessarily requiring extensive technical knowledge or expertise.

The process of selecting and transforming relevant variables or features from raw data to improve the performance of machine learning models.

A machine learning approach that allows multiple devices to collaboratively train a model using their local data without sharing it, enhancing privacy and security.

A technique in AI where a model learns to make accurate predictions by training on a very small number of labeled examples, allowing it to generalize to new, unseen data quickly and efficiently

Reconfigurable hardware that can be programmed to perform various AI tasks, such as image processing and natural language processing.

The process of taking a pre-trained machine learning model and making small adjustments or additional training on a specific task to improve its performance for that task.

A biometric technology that uses unique patterns found on an individual's fingers to identify and verify their identity, often used in security systems, law enforcement, and personal devices like smartphones.

A computational method used to predict how objects respond to physical forces by breaking them down into smaller, manageable parts called finite elements, allowing engineers to analyze stress, strain, and deformation in complex structures.

The process of feeding input data through a neural network in a forward direction, where each layer processes the data using its own activation function and passes the output to the next layer, ultimately generating an output from the network.

A large-scale, pre-trained model that serves as a base for a wide range of tasks and applications, which can be fine-tuned for specific purposes.

A metric used to evaluate the quality of images generated by generative models, such as Generative Adversarial Networks (GANs), by measuring the similarity between the distribution of generated images and real images based on computer vision features extracted from the Inception v3 model.

An FSD processor is a special computer chip inside a car that helps it drive itself by quickly processing all the data from cameras and sensors.

A European Union law that aims to protect the personal data of individuals by setting strict guidelines for how businesses collect, store, and use personal information, ensuring transparency and consent from users.

A type of artificial intelligence that can create new content, such as text, images, or music, by learning patterns from existing data.

A business intelligence approach that leverages machine learning and AI to generate insights and predictions from large datasets, enabling organizations to make data-driven decisions and optimize operations more effectively.

Using AI to optimize and enhance content generation processes, improving efficiency and quality through automated techniques.

A type of artificial intelligence model that can generate human-like text by learning patterns and structures from vast amounts of text data before being fine-tuned for specific tasks, allowing it to produce coherent and contextually relevant text.

When an app or website uses AI to instantly create or change what you see on the screen based on what you need or ask for.

The process by which an artificial intelligence system deduces the intentions or objectives of an agent based on its observed behaviors, often using models that simulate human-like reasoning and planning.

A cloud-based service that allows users to rent powerful graphics processing units (GPUs) on-demand, enabling them to perform complex computing tasks without the need for expensive hardware or maintenance.

The method of systematically visiting and checking each vertex in a graph to explore its structure and connections, akin to navigating through a network of points linked by edges.

A specialized electronic component that accelerates the rendering of graphics and images on digital screens, making it essential for smooth visuals in videos, video games, and other graphics-intensive applications.

A graphics processor is a special computer chip that quickly handles images, videos, and visual effects so everything looks smooth on your screen.

Green AI is artificial intelligence designed to use less energy and resources, so it’s smarter for the planet as well as for people.

Grokking means understanding something so deeply and completely that it just clicks and feels second nature, like riding a bike or speaking your native language.

Guidelines or constraints put in place to ensure that artificial intelligence systems operate within specified ethical, legal, and safety boundaries.

Like fact-checking an AI to see where it makes things up or gives wrong answers so you know how much you can trust it.

A specific type of input designed to elicit a particular response from a large language model (LLM), often requiring a detailed and structured approach to guide the model's understanding and generation of the desired output.

The integration of artificial intelligence systems with hardware components to optimize performance and efficiency.

Artificial intelligence systems that operate independently of a user interface, focusing solely on processing and generating data through APIs or other programmatic interfaces, allowing for seamless integration into various applications and systems

A federal law that aims to protect the privacy and security of patients' medical records and ensure continuous health insurance coverage for individuals who change or lose their jobs by standardizing electronic transactions and promoting the use of electronic media for healthcare data transmission.

A type of encryption that allows data to be processed and analyzed without being decrypted, ensuring the data remains secure and private.

Horizontal AI is artificial intelligence designed to perform general tasks (like writing, analyzing data, or answering questions) across many industries, instead of being built for one specific job or sector.

Combines human intelligence with artificial intelligence, allowing them to work together and learn from each other to achieve better outcomes and enhance each other's strengths and weaknesses.

An intensive support period after the initial deployment of an AI system to ensure it functions smoothly and address any issues that arise.

A configuration parameter that is set prior to training a machine learning model and affects its learning process and performance.

A technology that enables computers to identify and classify objects, people, and other elements within images, much like humans do.

The integration of intelligent digital technologies into manufacturing and industrial processes, enabling automation, real-time data analysis, and seamless communication between machines and humans to improve efficiency and productivity.

The duration it takes for a machine learning model to make predictions on new data.

The process of collecting, organizing, storing, and providing information within a company or organization to ensure its accuracy, accessibility, and effective use for decision-making and operations.

The process of finding and retrieving relevant information from large collections of data, such as documents, images, or videos, by matching user queries with the content of these collections.

Any data or information created, received, or collected by an organization.

When an AI is trained to better follow human-written instructions so it gives answers in the way people expect.

The intangible value of an organization's employees, skills, knowledge, and training that can provide a competitive advantage and drive long-term business value.

An autonomous entity that acts to achieve goals using observation through sensors and consequent actuators.

The integration of artificial intelligence techniques, such as machine learning and deep learning, into control systems to enable them to adapt, learn, and make decisions autonomously, enhancing their efficiency and reliability.

A smart computer tool that helps people make better decisions by analyzing data, spotting patterns, and suggesting the best options.

A cutting-edge technology that leverages artificial intelligence (AI) and natural language processing (NLP) to provide personalized and contextually relevant assistance to users, allowing them to interact with devices through voice commands, text inputs, or gestures.

A network of devices, vehicles, appliances, and other objects that can collect and share data over the internet without human intervention, making them "smart" and capable of interacting with each other and with humans in various ways.

The process of assigning specific meanings to symbols and expressions in formal languages, such as natural language, programming languages, or data representations, enabling AI systems to understand and process information in a way that is meaningful to humans.

The ability of an artificial intelligence system to learn and improve its performance without relying on external rewards or incentives, driven by internal factors such as curiosity, exploration, creativity, and self-regulation.

An international standard that helps organizations protect their information by setting up a systematic approach to managing and securing their data and systems.

The combined hardware, software, networks, and processes that work together to store, manage, and deliver information for businesses or individuals.

The process of repeatedly refining and improving AI models, data, and problem definitions through cycles of experimentation, analysis, and refinement, ensuring continuous improvement and better performance over time.

A strategy where you build on the model's previous outputs to refine, expand, or dig deeper into the initial answer by creating follow-up prompts based on the model's responses, allowing for more accurate and comprehensive results.

A smart search process that keeps refining your query based on feedback until it finds the most relevant information.

A simple machine learning technique that makes predictions based on the majority class of its k nearest neighbors in a feature space.

Valuable information and expertise that an organization possesses, including data, documents, procedures, and employee know-how, which can be used to create value and achieve objectives.

An evaluation process that identifies and assesses the knowledge assets within an organization to ensure they are effectively used and managed.

The process of using technology to automatically gather, organize, and apply existing knowledge to solve problems or complete tasks, freeing up humans to focus on higher-level decision-making and creative work.

A centralized repository of information that provides quick access to specific data, answers, and solutions, helping users find answers on their own without needing to contact support agents.

An economic system where knowledge and intellectual capabilities are the primary drivers of growth, innovation, and productivity, relying less on physical inputs and natural resources, and more on the creation, dissemination, and utilization of intangible assets like information, technology, and intellectual property.

The process of designing and developing computer systems that incorporate human expertise and knowledge to solve complex problems, typically involving the integration of artificial intelligence techniques and symbolic structures to represent and reason with knowledge.

The continuous sharing and dissemination of information, skills, and expertise within an organization, enabling employees to learn from each other and adapt to changing circumstances effectively.

A network of interconnected information, where entities (like people, places, and things) are linked by their relationships, helping computers to understand and use this data more effectively.

The process of capturing and documenting valuable insights, experiences, and expertise from individuals within an organization to make it accessible for others.

The process of creating, sharing, using, and managing an organization's information and knowledge resources to enhance its efficiency and decision-making.

The process of keeping and maintaining the information, skills, and experiences gained over time, ensuring that valuable insights and expertise are preserved and can be used effectively in the future.

The process of searching for and extracting relevant information from a large collection of data or documents.

The process of exchanging information, skills, or expertise among individuals or groups to enhance understanding and improve outcomes.

The isolation or compartmentalization of information, expertise, or skills within an organization, leading to a lack of cross-functional collaboration, hindered communication, and inhibited learning

The ability of a model or system to apply knowledge or skills learned in one context to another, often across different domains or tasks, enhancing its versatility and effectiveness.

The practice of using visual representations, such as charts and graphs, to make complex information and data easier to understand and interpret.

An open-source framework that allows developers to combine large language models with external data and computation to build AI applications.

How computers or humans understand, interpret, and respond to written or spoken words.

An advanced artificial intelligence system that not only understands user queries but also takes action based on those queries, effectively transforming AI from a passive tool into an active collaborator capable of executing complex tasks autonomously.

Advanced AI systems that analyze massive amounts of data to understand complex ideas and generate intelligent responses, similar to how humans make connections between different concepts.

IT support for AI that helps teams manage, monitor, and improve big AI models (like ChatGPT) so they work safely, reliably, and efficiently in real-world apps.

A type of artificial intelligence that can understand and generate human-like text by being trained on vast amounts of written data.

A complex mathematical framework used to analyze vast amounts of data to make predictions or decisions in various fields like finance or science.

Super-smart math tools that analyze tons of data to help people and businesses make better decisions and predictions.

A method used to analyze the meaning of words and phrases by examining the relationships between them in large amounts of text.

The progression from basic, one-off interactions with AI systems to more sustained and contextually rich relationships, with the potential for AI companions to develop deeper emotional connections and understanding over time.

A process in natural language processing that reduces words to their base or dictionary form, known as the lemma, to improve text analysis, search queries, and machine learning applications by normalizing different inflected forms of the same word into a single, standardized form

A method of searching for information that looks for exact matches of keywords or phrases within a database, ignoring variations in spelling or grammar, and is useful for finding specific information quickly but can struggle with nuances in language

A technology that uses laser light to measure distances and create detailed 3D maps of objects and landscapes, kind of like radar but with lasers instead of radio waves.

A type of artificial intelligence that learns from past experiences and observations, allowing it to make predictions and decisions based on both past and present data, but it does not retain this information in its memory for long-term learning or recall.

The process by which large language models generate responses or predictions based on the input they receive, utilizing patterns learned during their training to produce human-like text.

A type of artificial intelligence where computers learn from data and improve their performance over time without being explicitly programmed.

The behind-the-scenes system that helps data scientists turn smart computer models into reliable, working tools that businesses can actually use every day.

A technology that uses computer algorithms to automatically convert text or speech from one language to another, enabling global communication and business without the need for human translators.

A technology that allows devices to automatically exchange information without human intervention, enabling machines to communicate with each other and with central systems over wired or wireless networks.

Away to model decision-making where an agent chooses actions that affect what happens next, with outcomes depending only on the current situation and not the past.

A measure of how far off predictions are from actual values, calculated by averaging the absolute differences between them, giving you a clear idea of prediction accuracy.

A measure of how close a set of predicted values is to the actual values, calculated by averaging the squares of the differences between each predicted value and its corresponding actual value.

A guide or prompt for prompts that helps users form the most suitable question for an AI, essentially asking the AI to suggest the best prompts to use for a given aim, much like asking a librarian for book recommendations.

Like a label or note that describes what a piece of data is, making it easier to find, understand, and organize—like the tags on a luggage bag telling you who it belongs to and where it’s going.

A software development approach where a large application is broken down into multiple, independent, and specialized services that communicate with each other using APIs, allowing for greater scalability, flexibility, and maintainability.

An AI technique where only a few specialized "mini-expert" models are activated at a time to solve a problem, making the system faster and smarter without using all its power at once.

A situation where a generative model, such as a Generative Adversarial Network (GAN), is only capable of producing a limited number of distinct outputs or modes, resulting in low diversity and repetition of similar images.

A universal translator that lets different AI models and tools talk to each other smoothly without extra setup.

A set of rules that helps AI remember and use past conversations or data to give more relevant and accurate responses.

The process of assessing the performance and accuracy of AI or ML models using metrics like accuracy, precision, and recall.

A software design approach where a single, self-contained unit, often a large program or application, is developed and managed as a single entity, rather than breaking it down into smaller, independent components or microservices.

A computational technique that uses random sampling to model the behavior of complex systems and estimate outcomes or probabilities in various scenarios.

A deep linguistic analysis method that identifies the part-of-speech, lexical properties, and grammar of each token, essentially breaking down words into their smallest components to understand their meaning and structure.

The process of breaking down words into their smallest meaningful units, called morphemes, to better understand the structure and meaning of language, which is crucial for various natural language processing tasks such as machine translation, sentiment analysis, and text comprehension.

The process of breaking down words into their smallest meaningful parts, called morphemes, to understand how they are structured and how they relate to each other to convey meaning.

A system where multiple autonomous AI agents work together, communicating and collaborating to solve complex tasks more efficiently than any single agent could on its own.

A security process that requires a user to provide multiple forms of verification, such as a password, fingerprint, or one-time code, to ensure that only authorized individuals can access a system or account.

A technique in AI where the model looks at different parts of hidden (latent) information from multiple angles at once to better understand complex patterns or relationships.

A type of artificial intelligence that combines multiple types of data, such as text, images, audio, and video, to create more accurate and comprehensive insights by mimicking the way humans process information from different senses

When a program splits a repetitive task into several smaller parts and runs them at the same time on different "workers" (threads) so it finishes faster.

A process in natural language processing (NLP) that identifies and categorizes specific entities in text, such as names, locations, organizations, and dates, into predefined categories to extract structured information from unstructured text.

A NAND memory drive is a type of storage, like a USB stick or SSD, that saves data on flash memory chips instead of spinning disks, making it faster and more durable.

The process of using machines to automatically create human-understandable text from input data, such as prompts, tables, or images, aiming to produce text that is indistinguishable from that written by humans.

A technology that enables computers to understand, interpret, and generate human language, allowing them to interact with humans more naturally and efficiently

Enables users to ask questions or type queries in everyday language—like talking to a person—instead of using specific keywords, making it easier to find information.

When a computer smoothly changes from one task or topic to another using human-like language, so it feels more natural in conversation.

The ability of computers to comprehend and interpret human language, allowing them to understand and respond to natural language inputs like we do, making it a crucial technology for applications like chatbots, virtual assistants, and language translation tools.

Computational techniques inspired by the structure and function of the human brain, used to model and solve complex problems in machine learning and artificial intelligence.

A type of artificial intelligence that mimics the human brain's structure to process and learn from data, helping computers recognize patterns and make decisions.

A brain-computer interface (BCI) technology developed by Elon Musk's company, which aims to enhance human intelligence by implanting a chip in the brain, allowing people to control devices with their thoughts and potentially treating conditions like paralysis and blindness.

A type of artificial intelligence that combines the pattern recognition abilities of neural networks with the logical reasoning of symbolic AI to enhance understanding and decision-making.

A way of teaching AI by using evolution-like processes—like mutation and natural selection—to gradually improve the structure and performance of neural networks.

Chips that are designed to mimic the brain's structure and function, using artificial neurons and synapses to process information more efficiently and adaptively than traditional computers.

A new way of designing computers that mimics the structure and function of the human brain, using artificial neurons and synapses to process information in a more efficient and adaptable manner.

A set of guidelines and best practices that help organizations identify, protect, detect, respond to, and recover from cyber threats.

The system can produce different outcomes even with the same input, because it involves randomness, probabilities, or multiple valid paths to a solution.

A process in artificial intelligence (AI) that transforms data into a standard format to ensure all features are on the same scale, making it easier for AI models to analyze and learn from the data accurately.

Software or services that are hosted and managed within an organization's own infrastructure, typically on the company's own servers or data centers, rather than being hosted externally by a third-party provider.

A technology that allows users to quickly analyze and manipulate large amounts of data from multiple perspectives for business intelligence purposes.

Information that anyone can freely access, use, and share without restrictions.

A free toolbox and set of instructions that anyone can use to build and test robots, so people and companies don’t have to start from scratch.

Software or projects that are freely available for anyone to use, modify, and distribute, typically fostering collaboration and innovation within a community of developers and users.

A technology that lets computers read and convert printed or handwritten text in images or scanned documents into editable and searchable digital text.

The process of structuring and aligning an organization's people, roles, and processes to achieve its goals and strategy, ensuring it operates efficiently and effectively to achieve its objectives.

The process by which an organization improves itself over time by gaining experience, creating knowledge, and sharing that knowledge among its members to stay competitive and efficient.

The collective knowledge and information that an organization accumulates over time, encompassing both documented data and the personal experiences of its members, which can be used to improve decision-making and learning within the organization.

Only pay when you actually get the promised results, not just for the effort or service.

A situation where a machine learning model becomes too specialized to the specific training data it was trained on, making it unable to accurately generalize to new, unseen data and resulting in poor performance on new predictions

A thought experiment where an artificial intelligence is programmed to maximize the production of paperclips, leading it to pursue increasingly abstract and complex strategies to achieve this goal, often resulting in unexpected and humorous outcomes.

A parameter refers to a specific numerical value or input used in a model to estimate the probability of a particular AI-related outcome, such as the likelihood of an AI catastrophe, and understanding the uncertainty associated with these parameters is crucial for making informed decisions about AI development and risk mitigation.

A process where computers automatically assign a specific grammatical category, such as noun, verb, adjective, or adverb, to each word in a sentence to better understand its meaning and context.

A security method that eliminates the need for passwords by using alternative authentication methods, such as biometric data, one-time codes, or smart cards, to verify a user's identity and grant access to digital systems.

The process of identifying and analyzing regularities or patterns in data to make sense of it and draw conclusions.

A set of security standards designed to protect sensitive cardholder data by ensuring that merchants and service providers maintain secure environments for storing, processing, and transmitting credit card information.

A simulated cyber attack on a computer system or network to identify vulnerabilities and weaknesses, helping to strengthen security measures and prevent real-world breaches.

A set of fundamental concepts in machine learning that are used to build and train neural networks, which are the core components of many AI systems.

Any data that can be used to identify a specific person, such as their name, address, phone number, date of birth, or other personal details, which can be used to distinguish them from others and potentially compromise their privacy

A type of cybercrime where attackers use fraudulent emails, texts, or messages to trick people into revealing sensitive information, such as passwords or financial details, by pretending to be a legitimate source.

Artificial intelligence that is integrated into physical systems, like robots or smart devices, enabling them to sense, learn, and interact with the real world autonomously.

A small-scale test or trial run of a new AI system or feature to ensure it works as intended before full-scale deployment.

Areinforcement learning method where an AI learns the best way to act by directly adjusting its decision-making rules (the "policy") based on how much reward it gets.

A way for AI models to keep track of the order of words in a sentence so they understand meaning beyond just the words themselves.

A robot that’s already read a million books so it can help you faster—without having to start learning from scratch every time.

The use of historical data, statistical algorithms, and machine learning techniques to forecast future outcomes and trends.

A strategy that uses data analysis and sensors to predict when equipment will need maintenance, helping to prevent unexpected failures and reduce downtime.

A statistical technique used to create a model that can predict future outcomes based on historical data.

The use of data, algorithms, and machine learning to recommend actions that can help achieve desired outcomes or solve specific problems.

A dedicated cloud computing environment for a single company, where the infrastructure is controlled and managed by the organization itself, offering enhanced security, scalability, and customization compared to public cloud services.

Using technology to perform repetitive tasks automatically, saving time and reducing the need for manual effort.

When your product solves a real problem so well that customers keep coming back and even tell their friends about it.

When a product that was once in high demand suddenly loses its appeal, causing a sharp decline in sales or user engagement.

Where your website or application is live and accessible to the public, meaning it's the final stage where everything is set up and running for users to interact with it.

A fancy way physicists use to describe super complex systems—like a bunch of particles interacting - in a neat, compressed format so computers can understand and simulate them more easily.

The suggestion or question you enter into an AI chatbot to get a response.

The ability of AI to use information from previous interactions to color future responses

Crafting effective prompts or input instructions for AI systems to generate desired outputs or responses, enhancing their performance and accuracy in various tasks.

A technique where you adjust the way you ask a language model questions to get more accurate or relevant answers.

A small-scale test or demonstration to prove the feasibility and potential of an idea or product before investing more time and resources into its development.

Information that a company owns and protects because it gives them a competitive advantage, like secret recipes, customer lists, or unique research.

A type of machine learning algorithm that helps an agent learn to make the best decisions in a given situation by interacting with the environment and receiving rewards or penalties for its actions, without needing a detailed model of the environment.

Gathering and analyzing non-numerical data, such as opinions, experiences, and behaviors, to gain a deeper understanding of a topic or issue.

Using numerical data and statistical methods to analyze and understand phenomena, often aiming to identify patterns, trends, and correlations.

A type of computing that utilizes the principles of quantum mechanics to perform complex calculations much faster than traditional computers.

The process of crafting a search query or request for information in a structured manner to retrieve relevant data from a database or search engine.

The process of improving the performance and efficiency of database queries by selecting the most optimal execution plan to retrieve data quickly and accurately.

RAM (Random Access Memory) is like your computer’s short-term memory—it temporarily stores data your device is actively using so it can work faster, but it gets cleared when the power is off.

A technique used in large language models that involves generating prompts to elicit specific responses, similar to how a programmer writes code to achieve a desired outcome.

A type of artificial intelligence that can only respond to the current input and does not have any memory or ability to learn from past experiences, making it highly specialized and effective in specific tasks like playing chess or recognizing patterns in data.

A way for AI to think through a problem step by step and take actions—like searching or calculating—to figure out the best answer, just like a person would.

A system that uses data and algorithms to suggest products, services, or content to a user based on their past behaviors, preferences, and similarities to other users, aiming to provide a personalized and relevant experience.

A type of artificial neural network that can learn patterns in data over time, making it useful for tasks like speech recognition, language translation, and predicting future events.

Simulating adversarial attacks to identify vulnerabilities and enhance the robustness of AI systems against potential threats.

A type of machine learning technique used to predict continuous numerical values based on input features, such as predicting house prices based on factors like size, location, and number of bedrooms.

A type of machine learning where an agent learns to make decisions by trial and error, receiving feedback in the form of rewards or penalties based on its actions.

A machine learning technique that uses human feedback to train AI agents to perform tasks by rewarding them for actions that align with human preferences, making them more effective and efficient in achieving their goals.

A reinforcement signal is like giving a "thumbs up" or "thumbs down" to an AI so it learns which actions are good and which ones to avoid.

A structured system for organizing and storing data in tables with relationships between them, making it easier to manage and retrieve information.

A specific instruction or command given to an artificial intelligence system to perform a particular task or function, such as processing data, making decisions, or generating output.

The process of generating appropriate and contextually relevant responses in conversational systems such as chatbots or virtual assistants.

The practice of designing, developing, and deploying artificial intelligence systems in a way that ensures fairness, transparency, and accountability, and minimizes harm.

A type of web service that allows different software applications to communicate and interact over the internet using standard HTTP methods like GET, POST, PUT, and DELETE.

An advanced natural language processing technique that combines real-time data retrieval with response generation, allowing AI models to dynamically fetch and integrate external information while formulating answers, rather than relying on a single retrieval step before generating a response.

LLM using additional context, such as a set of company documents or web content, to augment its base model when responding to prompts.

Teaching a robot to get better at its job by making small adjustments based on how it performs, so it works smarter and more accurately over time.

Powerful AI systems trained on lots of data to help robots understand and interact with the world more intelligently and flexibly, like learning to see, move, and make decisions in different situations.

Powerful AI brains that help robots learn many different tasks—like seeing, moving, or following voice commands—so they can work more like humans in everyday situations.

A technology that uses software robots to automate repetitive, rule-based tasks typically performed by humans, improving efficiency and accuracy.

Role augmentation is the use of technology, like AI, to enhance and support a person's job tasks, making them more efficient and effective without replacing their role.

The ability of AI systems to ask questions and seek understanding, often mimicking human curiosity and self-awareness, which can lead to more complex and nuanced interactions with humans and other AI systems.

Away for AI models to decide which parts of the input (like words in a sentence) should pay the most attention to each other in order to understand the meaning.

A type of artificial intelligence that possesses a sense of self, understanding its own state and existence, and can reflect on its actions, learn from experiences, and adapt its behavior accordingly.

The ability of a language model to provide consistent and logical responses to questions or scenarios, ensuring that its answers align with its understanding of the world and do not contradict each other.

When an AI teaches itself by finding patterns in unlabeled data, kind of like solving puzzles it creates on its own to get smarter without needing a teacher.

A process that helps computers understand the meaning and context of human language by analyzing the relationships between words and phrases, allowing them to extract insights and make decisions based on the text

An open-source software development kit (SDK) that allows developers to easily integrate artificial intelligence (AI) models, such as large language models, with conventional programming languages like C# and Python, enabling the creation of AI-powered applications.

A process in natural language processing that assigns labels to words or phrases in a sentence to indicate their roles in the sentence, such as agent, goal, or result, to help machines understand the meaning of the sentence

A way for computers to understand the meaning behind your search query, giving you more accurate and relevant results by considering the context and intent behind your search, rather than just matching keywords

The process of labeling content with meaningful tags that describe its context, making it easier for AI and search engines to understand and organize information.

The process of using natural language processing and machine learning techniques to determine the sentiment or emotional tone expressed in text, such as positive, negative, or neutral.

The automated process of identifying and categorizing the emotional tone expressed in text or speech, such as positive, negative, or neutral sentiments.

A method where a series of prompts are used in a specific order to elicit a desired response from a language model, often involving a sequence of questions or tasks that build upon each other to achieve a particular goal or understanding.

A cloud computing model where the cloud provider automatically manages the infrastructure, allowing developers to run code without worrying about server management, scaling, or maintenance.

A compliance framework that ensures a service organization's internal controls are effective in handling and reporting financial data securely and accurately, providing assurance to users that their financial information is properly managed.

A security framework that ensures organizations protect customer data by implementing robust controls and policies, similar to how you would protect your personal belongings by locking your doors and keeping valuables secure.

When a program runs one task at a time in a continuous cycle, finishing each before starting the next, all on a single processing thread.

The difference between the skills and knowledge that workers currently possess and the skills and knowledge that employers need to remain competitive in the modern workforce.

Relatively small, specific, and actionable datasets that are often used to inform immediate business decisions, as opposed to large, complex datasets that require advanced analytics and processing.

A type of artificial intelligence that can understand and generate human-like text, but is typically less complex and less powerful than larger models, making it suitable for specific tasks or applications where a more focused and efficient model is needed

A municipality that uses information and communication technologies (ICT) to increase operational efficiency, share information with the public, and improve both the quality of government services and citizen welfare.

A way to organize data in a database where related information is broken into smaller linked tables to reduce repetition and keep things tidy.

A flexible and adaptable piece of text that is used to guide a language model to perform a specific task, often by being prepended to the input sequence to help the model understand the task better.

A set of tools and resources that helps developers build apps or software for a specific platform or system.

A structured process that outlines the stages involved in creating software, from planning and analysis to design, implementation, testing, and maintenance, ensuring a well-organized and efficient approach to software development.

The recipe for a computer program—written in a language humans can read so the computer knows exactly what to do.

A technique in AI that makes models faster and more efficient by focusing only on the most important parts of the data instead of paying attention to everything at once.

A technology that enables the interaction between digital content and the physical world, allowing users to seamlessly blend virtual elements with their real-life environment.

Hardware designed specifically for AI tasks, such as AI-specific processors and AI-specific memory architectures.

The ability of a computer to understand and transcribe spoken language into text, allowing for hands-free interaction with devices and applications.

A test space that mimics the real production environment, allowing developers to thoroughly check and refine software before it's released to the public.

A simple way to organize data for reporting, where one big table of numbers (like sales) connects to smaller tables of details (like products, customers, and dates), kind of like a star shape.

A process in natural language processing that reduces words to their root form by removing suffixes and prefixes, allowing for more effective text analysis and comparison

A large language model that can generate human-like text but lacks true understanding of the meaning behind the words, essentially mimicking patterns without comprehension.

A method of annotating scholarly articles with specific classes, such as background, methods, results, and conclusions, to create a machine-readable summary that can be used for more effective search and analysis of the article's content

Organized and well-formatted information that is typically stored in databases or spreadsheets, making it easy to search, analyze, and process.

An AI technique that allows you to take an image and transform it into a new image with a different style, such as a painting or a cartoon, while keeping the original content intact, creating a unique and artistic visual effect.

A business model where you pay a recurring fee—monthly, yearly, or on another schedule—to continuously access a product or service instead of buying it outright.

A hypothetical form of AI that surpasses human intelligence by developing its own thinking skills and cognitive abilities, allowing it to perform tasks that are beyond human capabilities.

A type of artificial intelligence where models are trained on labeled data, enabling them to make predictions or decisions based on input-output pairs provided during training.

A supervised learning algorithm used for classification and regression tasks, particularly effective in high-dimensional spaces.

The collective behavior of a group of simple individuals, like ants or bees, working together to achieve complex tasks without a central leader.

The use of symbolic representations, such as rules and logical expressions, to reason and solve problems, which is distinct from the connectionist approach of deep learning and neural networks.

The ability of a computer to understand and analyze human language by identifying and grouping words with similar meanings, which helps improve the accuracy and efficiency of language-based applications such as search engines and language translation systems

Information that is artificially manufactured rather than generated by real-world events.

Any image, video, audio, or text created or altered by artificial intelligence instead of being captured or recorded from the real world.

Information that is organized in a table format, making it easier to read, analyze, and interpret by displaying values in rows and columns.

The understanding and skills people have gained through personal experience and context, which is often difficult to articulate or document.

Task automation is using technology to handle repetitive tasks automatically, saving time and reducing the need for manual effort.

The practice of taking shortcuts or making suboptimal design or implementation decisions to expedite development, which can lead to increased complexity, maintenance costs, and difficulties in the long run, similar to taking out a loan to buy something now and paying interest later.

A hypothetical future event where artificial intelligence surpasses human intelligence, leading to exponential growth and potentially uncontrollable technological advancements that could fundamentally change human civilization beyond recognition.

The physical environmental temperature that can affect human performance and cognitive abilities, which is relevant to AI research as it can influence how humans interact with AI systems and how AI systems are designed to adapt to different environmental conditions.

Custom-designed AI accelerators developed by Google to optimize machine learning workloads.

The process of transforming raw, unstructured text data into a structured format that can be understood by machines, involving steps such as cleaning, tokenization, normalization, and encoding to prepare the text for analysis and machine learning tasks.

A technology that uses artificial intelligence to create images from natural language descriptions, allowing computers to generate realistic images based on text inputs like sentences or paragraphs.

The ability of artificial intelligence to understand and model the thoughts, intentions, and emotions of other agents, such as humans or other artificial intelligences, enabling more nuanced social interactions and effective communication.

The amount of time it takes for an idea or product to go from concept to being available for customers to buy or use.

The sweet spot where your product is launched at the right time to meet strong customer demand and market readiness, maximizing its chances of success.

How long it takes after you start using a product or service before you actually see real benefits or results from it.

A token is a unit of text, such as a word or a part of a word, that is used as a basic element for processing and analyzing language.

The process of breaking down text into smaller pieces, such as words or phrases, to make it easier for computers to understand and analyze.

A way to analyze large amounts of text data to identify and group related ideas or themes, like topics, within the content.

The set of data used to fit and train a machine learning model, which is then used to make predictions or classify new, unseen data.

A machine learning technique where a model developed for one task is reused as the starting point for a model on a second task.

A type of deep learning model in AI that learns context and meaning by tracking relationships in sequential data, such as words in a sentence, allowing it to understand and generate human-like text with unprecedented accuracy.

A method of evaluating a machine's ability to exhibit intelligent behavior indistinguishable from that of a human, typically through conversation.

Information that lacks a predefined data model or organization, such as text documents, images, videos, or social media posts, making it challenging to analyze using traditional methods.

A type of artificial intelligence where models analyze and find patterns in unlabeled data without explicit guidance, allowing them to discover hidden structures and relationships on their own.

Like paying for electricity or water—you only pay for what you actually use, instead of a fixed monthly fee.

The screen, buttons, menus, and other elements you see and use to interact with a device, app, or website.

The process of gathering information about people's needs, behaviors, and experiences to design products, services, or experiences that meet their expectations.

A type of deep learning model that compresses data into a lower-dimensional space and then reconstructs it, allowing it to generate new data that resembles the original data while also performing tasks like dimensionality reduction and anomaly detection.

Smart quantum programs that team up with regular computers to find the best answers to really tough problems by tweaking settings over and over.

A type of database optimized for storing and querying spatial data, such as geographic information system (GIS) data, allowing for efficient management and analysis of location-based information.

A method that uses mathematical vectors to represent and efficiently search through complex, unstructured data, allowing for more accurate and contextually-aware searches by comparing the similarity between query vectors and stored data vectors.

Vertical AI is artificial intelligence built for one specific industry or niche (like healthcare, finance, or retail) instead of trying to do everything for everyone.

When you tell an AI what you want your software to do in plain English, and it writes the code for you.

A way of using AI to turn your mood or style ideas—like “calm and cozy” or “bold and exciting”—into actual designs for websites, apps, or products.

A software program that can perform tasks or provide information for users through conversation, typically using voice commands or text interactions.

A secure and private connection between your device and the internet, allowing you to browse anonymously and access geo-restricted content by encrypting your data and routing it through a remote server.

A technique used in speech recognition and animation where the movements of a speaker's mouth and lips are matched to specific sounds or phonemes (like "ah" or "oh") to create a more realistic and natural-looking lip sync in videos or animations.

A technology that enables computers to understand and process spoken language, allowing users to interact with devices and applications using their voice.

A type of artificial intelligence that is focused on a particular task and can't learn beyond its skill set.

A program that automatically browses the internet to index and collect information from websites for search engines and other applications.

A way for applications to communicate with each other in real-time by sending HTTP requests to a specific URL when a specific event occurs, allowing for instant updates and notifications.

The next iteration of the internet, which aims to decentralize the web by giving users more control and ownership through blockchain technology, cryptocurrencies, and non-fungible tokens (NFTs), allowing them to participate in the governance and decision-making processes of online platforms and services.

The use of technology to streamline and automate repetitive tasks and processes, improving efficiency and reducing the need for manual intervention.

A technique in machine learning where a model can recognize and classify new concepts without any labeled examples, using pre-trained knowledge and auxiliary information to bridge the gap between known and unknown classes

Zero-Trust Architecture is a security approach that treats every user, device, and AI system as untrusted by default, requiring continuous verification before granting access to data or applications.