AI & Technical Terms Glossary

Find definitions for artificial intelligence and technical terms used across our website.

Last updated: June 15, 2024

A

AGI (Artificial General Intelligence)
Hypothetical AI with the ability to understand, learn, and apply knowledge across a wide range of tasks at a level equal to or exceeding human capabilities.
Related terms:
AI (Artificial Intelligence)
Computer systems that can perform tasks requiring human intelligence, such as visual perception, speech recognition, decision-making, and translation between languages.
Related terms:
AI Alignment
The challenge of ensuring AI systems act in accordance with human intentions, values, and preferences, even as they become increasingly autonomous.
Related terms:
AI Applications
Practical implementations of AI technology to solve specific business or consumer problems.
Related terms:
AI Ethics
Field concerned with ensuring AI systems are designed and deployed in ways that align with human values and ethical principles.
Related terms:
AI Governance
Structures, processes, and policies organizations implement to manage AI development and deployment responsibly.
Related terms:
AI Infrastructure
Hardware, software, and networking components that support the development, training, and deployment of AI systems at scale.
Related terms:
AI Reasoning
Capability of AI systems to process information logically, draw inferences, and reach conclusions based on available data and knowledge.
Related terms:
AI Safety
Research field focused on ensuring advanced AI systems remain beneficial, controllable, and aligned with human values as they become more capable.
Related terms:
Attention Mechanism
Neural network component that allows models to focus on different parts of the input when producing outputs, crucial for understanding context in language.
Related terms:

B

Business Automation
Use of technology to execute recurring tasks or processes where manual effort can be replaced with automated systems.
Related terms:

C

Chain of Thought
Prompting technique that encourages AI models to show their reasoning process step-by-step, leading to more accurate and explainable results.
Related terms:
Clustering
Unsupervised learning technique that groups similar data points based on their features without prior labeling.
Related terms:
Computer Vision
Field of AI focused on enabling computers to interpret and understand visual information from the world, such as images and videos.
Related terms:

D

Data Science
Interdisciplinary field using scientific methods, processes, algorithms and systems to extract knowledge from structured and unstructured data.
Related terms:
Deep Learning
A machine learning technique based on artificial neural networks with multiple layers that progressively extract higher-level features from raw input.
Related terms:
Diffusion Models
Type of generative AI model that creates images by gradually removing noise from random data, powering tools like DALL-E, Midjourney, and Stable Diffusion.
Related terms:
Distributed Computing
System where components located on networked computers communicate and coordinate actions to appear as a single coherent system, often used in large-scale AI.
Related terms:

E

Embeddings
Dense numerical vector representations that capture semantic meaning of text, images, or other data, allowing machines to understand similarities between concepts.
Related terms:

F

Federated Learning
ML approach that trains algorithms across multiple devices or servers holding local data samples, without exchanging the data itself, enhancing privacy.
Related terms:
Fine-tuning
Process of further training a pre-trained model on a specific dataset to adapt it for particular tasks or domains, improving performance for specialized applications.
Related terms:

G

Generative AI
AI systems that can generate new content including text, images, audio, code, or other media in response to prompts.
Related terms:

I

Image Recognition
AI technology that identifies objects, people, scenes, and actions in images, enabling applications from facial recognition to medical diagnostics.
Related terms:

K

Knowledge Base
Organized repository of information that AI systems can access to retrieve facts, rules, and domain-specific knowledge when generating responses.
Related terms:

L

LLM (Large Language Model)
Advanced AI systems trained on vast text datasets that can understand and generate human-like text, code, and other content.
Related terms:

M

Machine Learning
A subset of AI focused on building systems that learn from data rather than following explicit programming instructions.
Related terms:
Multimodal AI
AI systems capable of processing and generating multiple types of data (text, images, audio, video) simultaneously, understanding connections between different modalities.
Related terms:

N

Narrow AI
AI systems designed to perform specific tasks within a limited domain, as opposed to general intelligence across multiple domains.
Related terms:
Neural Networks
Computing systems inspired by the biological neural networks of animal brains, designed to recognize patterns and solve complex problems.
Related terms:
NLP (Natural Language Processing)
Technology that enables computers to understand, interpret, and generate human language in useful ways.
Related terms:

P

Pre-training
Initial phase of model training on large general datasets before fine-tuning, allowing models to learn broad language patterns and knowledge.
Related terms:
Privacy-Preserving AI
AI technologies and methods designed to protect personal data while still delivering useful insights and functionality.
Related terms:
Process Efficiency
Measure of how effectively resources are used to achieve desired outcomes in business operations.
Related terms:
Prompt Engineering
Practice of designing and optimizing input prompts to AI systems to elicit desired outputs, improving the quality and relevance of generated content.
Related terms:

R

RAG (Retrieval-Augmented Generation)
A technique that enhances LLM outputs by retrieving relevant information from external knowledge sources before generating responses, improving accuracy and reducing hallucinations.
Related terms:
Responsible AI
Framework for developing AI systems that are ethical, transparent, fair, and accountable throughout their lifecycle.
Related terms:

S

Sentiment Analysis
Use of NLP to identify and extract subjective information from text, determining whether the writer's attitude is positive, negative, or neutral.
Related terms:
Superintelligence
Theoretical form of AI that would surpass human cognitive capabilities across virtually all domains, potentially leading to rapid self-improvement.
Related terms:
Supervised Learning
ML technique where the model is trained on labeled data, learning to map inputs to outputs based on example input-output pairs.
Related terms:

T

Text Analytics
Process of deriving meaningful information from text data using various techniques including statistical analysis and ML.
Related terms:
Training Data
The dataset used to train an ML model, containing examples the system learns from to make predictions or decisions.
Related terms:
Transfer Learning
ML technique where knowledge gained from training on one task is applied to a different but related task, reducing the need for extensive new training data.
Related terms:
Transformer
Neural network architecture that uses self-attention mechanisms to process sequential data, forming the foundation of modern LLMs like GPT and BERT.
Related terms:

U

Unsupervised Learning
ML technique where the model learns patterns from unlabeled data without explicit guidance or target outcomes.
Related terms:

V

Vector Database
Specialized database that stores data as high-dimensional vectors, enabling efficient similarity searches crucial for RAG systems and other AI applications.
Related terms:

W

Workflow Optimization
Process of improving efficiency by streamlining operations, reducing bottlenecks, and automating repetitive tasks.
Related terms: