Skip to content
Greg Lum

Learn

Learn AI

Get your bearings in AI: what familiar tools do, how answers are made, and what deserves a closer look. Start with everyday questions or explore how the systems are built.

Making Sense of AI

What AI means, how familiar tools work, and how to judge the help they give. No technical background or AI account needed. Read in order, or start with the question you came with.

Start reading

  1. What do people mean by AI?

    See how predicting, recognizing, recommending, and creating content fit under the name AI.

  2. Why is everyone talking about AI now?

    Follow the changes that brought AI from learning patterns to tools you can talk to.

  3. ChatGPT, Claude, and All Those Names

    Untangle the company, model, assistant, and tool behind familiar AI names, with no technical background needed.

  4. How does an AI answer get made?

    Follow an event notice into an AI answer, and check what context, generation, and search each contribute.

  5. Getting Useful Help from AI

    Turn a vague request into a task an assistant can work on, then check and improve its reply.

  6. How can I check an AI answer?

    Check the facts, open the sources, and notice what is missing. A confident answer or a citation can still be wrong.

  7. What should I share, and what can it do?

    Check saved chats, training choices, and connected tools, and see why drafting and sending need different permissions.

  8. Making Sense of What You Hear About AI

    Turn an AI headline into a task you can assess, and separate results from predictions.

Find Your Path

Four routes through the guides and a few working pages. They overlap, so follow more than one or switch partway.

Explore by Topic

Every guide, grouped by subject. Some guides use code and project examples; you can read the explanations without running them.

AI Foundations

What AI is, where the names come from, and how an answer gets made.

Working with AI

Asking for useful help, protecting information, and checking what comes back.

Models and Decisions

When a job needs writing, finding, ranking, a bounded choice, or plain code.

  • Writing an Answer or Choosing One

    Follow a bike shop request through writing, finding, ranking, choosing, and the jobs for ordinary software.

    · 5 min read

  • Finding Before Answering

    Follow a holiday returns question from search to reply, see why a better choice cannot recover a missing policy, and check each stage.

    · 6 min read

  • Should a Model Make This Decision?

    Compare code, classifiers, and models for a fixed choice, set an error limit, and test on fresh cases.

    · 5 min read

  • Choosing the Cheapest Model That Passes

    Write the bar first, test the smallest model on frozen cases, and move up one size only when a check fails.

    · 5 min read

  • What makes an AI request cost more?

    Work out the token bill for a book-swap question box, then compare caching, batching, and sending less.

    · 6 min read

  • Where should an AI model run?

    Compare who sees your text, what you pay, and who maintains the model when you use a hosted service, a cloud platform, or your own machine.

    · 6 min read

Building and Evaluating

Designing software around a model, testing it, and reading the evidence for a claim.

  • Designing the System Around a Model

    Choose the simplest AI design that does the job, then decide its context, limits, recovery, and evidence, with Ask the Archive as the example.

    · 10 min read

  • How to Read a Model Claim

    Ask 7 questions about a model comparison: its baseline, version, sample size, total cost, and who ran it.

    · 6 min read

  • One Rulebook for Two Agents

    Give Claude Code and Codex the same user-level rules through an import and a symbolic link.

    · 3 min read

  • From a Public Repo to a Better Skill

    Find advice your skill needs, check its source and license, and test the change against a known failure.

    · 3 min read

  • Catching What Agents Miss

    See how fresh review, browser measurements, and accessibility tests caught 4 defects in a Draftroom change that passed local checks.

    · 3 min read

Risks, Rules, and Security

How AI systems get misused, and what the rules require and when.

AI in the World

What AI changes at work and across fields, and what it takes to run.

  • Will AI take my job, or change it?

    Find out what studies measure about AI and work, then test which tasks an assistant could help you finish and check.

    · 6 min read

  • How is AI used in different fields?

    Follow documented AI uses across 9 fields: what goes in, who checks the result, and what a mistake costs.

    · 7 min read

  • How much energy does AI use?

    Check what electricity and water estimates count, from a single AI question to training a model and running data centers.

    · 6 min read

Try an Example

Pages on this site where you can see the guides' ideas at work.

Agent Skills

Skill descriptions for Claude Code and Codex, with their upstream sources and licenses.

Browse the catalog

Learning Log

  1. Prepared my skills repository for public release

    A cleaned export of the catalog, with an installer dry run, a validator, and license notices.

    Read more : Prepared my skills repository for public release

  2. Studied public motion skills for the film skill

    Adopted timing defaults, script checks, and a critique loop into motion-film, with each source and license recorded.

    Read more : Studied public motion skills for the film skill

  3. Cleaned up my development environment

    Retired Zed, Neovim, Warp, Podman, Caveman, an offline semantic search tool, and a local docs index. Context7 replaced the docs index.

    Read more : Cleaned up my development environment

  4. Wrote down how I split work between agents

    Claude Code mainly implements. Codex mainly refines writing, researches, and configures the machine. It is a habit, not a rule.

Full log Follow in a feed reader

Keep Learning

Courses and explanations from other people, with a different way into the subject.

  • University of Helsinki and MinnaLearn

    Elements of AI

    A broader introduction to AI, with explanations and exercises you can work through at your own pace.

    Free registration gives full course access; check the provider for completion options.

  • 3Blue1Brown

    Large language models explained briefly

    See how a language model builds an answer, one piece at a time.

    Public article and embedded video; no course registration

  • The Open University / OpenLearn · Microsoft collaboration

    An Introduction to Artificial Intelligence

    Explore how AI developed and the questions it raises for everyday life and society.

    Free public reading; account needed for all activities and completion features

  • OpenAI Academy · Product guide

    ChatGPT Fundamentals

    Get a practical introduction to asking ChatGPT for help and improving an answer through follow-up questions.

    Public guide; trying ChatGPT has separate account and product conditions

Browse all resources