Artificial Intelligence

AI Basics

  1. Home
  2. AI Basics
How AI Works

AI learns, predicts and generates content.

  • AI learns by studying examples (data) to recognize patterns, much like how people learn by exploring concepts, practicing and observing the world around them.
  • AI predicts by using what it’s learned to answer questions, provide suggestions or even create new content.
  • AI generates new content (text, images, code, etc.).

Simply put, AI doesn’t "think" like humans—it learns from data, predicts what makes sense based on its training and generates new content by identifying patterns in vast amounts of information.

How AI Works image

What is Artificial Intelligence?

Duration: 9:26

Presenter: Dr. Michael Littman, U.S. National Science Foundation

This video explores the early development of artificial intelligence and highlights humanity’s ongoing role in shaping its future. Learn how AI began, how it's evolving and what that means for us today.

AI Terms and Definitions
What AI Can & Can’t Do

What AI Can Do

  • Speed up workflows.
  • Generate ideas.
  • Enhance decision-making.
  • Automate repetitive tasks.
  • Identify trends and insights.
  • Comparative analysis.

What AI Can’t Do

  • Think like humans; it predicts outcomes based on data.
  • Guarantee accuracy; it may generate biased or false info.
  • Be trusted to operate without human oversight.
  • Create with originality or emotional depth.
  • Make ethical decisions or understand morality.
  • Possess self-awareness or consciousness.
Differences in AI
AI Hierarchy. AI, ML, Deep Learning, GenAI, LLMS. Chatbots and AI Assistants like Copilot and NotebookLM.

Types of AI

  • Machine Learning (ML): A way to build AI by using algorithms that learn patterns from data. ML systems improve with more data.
  • Deep Learning: A type of machine learning that uses layered systems (neural networks) to recognize complex patterns in data. It powers GenAI tools that create content like text, images or audio.
  • Generative AI (GenAI): AI models that generate new content (like text, images or audio) by learning from large datasets. Open-sourced models (Llama 2, Stable Diffusion) allow developers to modify them, while closed-source models (ChatGPT, Gemini) restrict access to their parameters and training data.
  • Large Language Models (LLMs): Advanced natural language processing (NLP) systems trained on massive text datasets, used for generating and analyzing content.
  • Chatbots: Interactive systems that provide automated responses or assistance using conversational text.

AI, Machine Learning, Deep Learning and Generative AI Explained (10:00) Learn more about the distinctions between AI, ML, DL and foundation models and how these technologies have evolved (2024).

What are Generative AI models? (8:46) Understand more about a popular form of generative AI, large language models and how they function (2023).

Practical Uses for GenAI

AI tools come in many forms, each designed to streamline tasks, boost creativity and improve efficiency. Here are some common types and how they can help. AI that...

Understands and works with text
  • Enhances reading, writing and organization of information.
  • Summarizes long documents or articles.
  • Suggests better ways to word emails or papers.
  • Answers questions or explains complex topics.
Creates images and designs
  • Simplifies visual design without requiring advanced skills.
  • Creates posters, graphics and social media images.
  • Generates custom visuals for presentations or projects.
  • Edits and enhances photos automatically.
Analyzes data and finds patterns
  • Extracts insights from large datasets to support decision-making.
  • Identifies trends and correlations.
  • Makes data-driven predictions.
  • Enhances research and problem-solving.
Sees and understands pictures and videos
  • Interprets and organizes visual content for easier access.
  • Identifies objects, scenes and text in images.
  • Sorts and categorizes photos and videos.
  • Scans and converts handwritten notes into digital text.
Makes recommendations
  • Provides personalized suggestions based on your needs.
  • Recommends relevant resources and tools.
  • Assists in selecting software and apps.
  • Offers solutions and ideas for problem-solving.
Risks Associated with AI
  1. Energy Consumption Training large AI models, especially deep learning models, requires significant computational power, leading to high energy consumption and environmental impact (Nature, 2024).
  2. Privacy Concerns AI technologies often involve the collection and analysis of large amounts of personal data, raising significant privacy and security issues (HAI, Stanford University, 2024).
  3. Job Displacement Automation driven by AI could lead to job losses in certain sectors, creating economic and social challenges (World Economic Forum, 2023).
  4. Security Risks AI can be exploited for malicious purposes, such as developing advanced cyberattacks or autonomous weapons (NIST, 2025).
  5. Catastrophic Risks As with all powerful technologies, advanced AI must be handled with great responsibility to manage the risks and harness its potential (Center for AI Safety, 2023).

What is AI Literacy?

Duration: 8:34

Presenter: Ben Jones, Data Literacy CEO

In this video, Data Literacy CEO Ben Jones explains what "AI Literacy" means and explores its four key elements: the ability to recognize, grasp, use and critically assess artificial intelligence and its impact on the world.

Additional Resources

What is AI

Bias in AI

Practical Uses