AI & Data Fundamentals

Units and bites

1. Unit 1·6 bites·About 18 min

   What is AI?

   No jargon: AI is software that learns from examples to make predictions.

   1. AI in one sentence

      Strip away the hype and AI is software that learns to make predictions from examples.

      3 min

   2. Rules vs examples

      Old software followed rules people wrote. AI finds its own patterns in examples.

      3 min

   3. AI, machine learning and generative AI

      Three words people use as synonyms, and how they actually nest.

      3 min

   4. How a chatbot writes

      A large language model predicts the next piece of text, one piece at a time.

      3 min

   5. Brilliant at some tasks, clumsy at others

      AI's abilities are uneven, so judge it task by task.

      3 min

   6. AI you already use

      Spam filters, arrival times and recommendations: AI has been around you for years.

      3 min
2. Unit 2·6 bites·About 18 min

   Prompts

   How to instruct AI so the answer is useful, and what you should never paste in.

   1. A prompt is a brief

      Write to AI the way you'd brief a smart new colleague who knows nothing about your situation.

      3 min

   2. Context, task, format

      A three-part checklist for almost any prompt.

      3 min

   3. Show an example

      One good example often beats a paragraph of description.

      3 min

   4. Say who it's for

      Naming the audience changes the vocabulary, the depth and the tone.

      3 min

   5. Refine, don't restart

      The first answer is a draft. Steer it with specific feedback.

      3 min

   6. What never goes in a prompt

      Personal data and company secrets need your company's rules first.

      3 min
3. Unit 3·8 bites·About 24 min

   Patterns, not magic

   How AI learns from data, and why bad data makes bad AI.

   1. Learning from examples

      Training data, labels, and why the examples have to look like the real world.

      3 min

   2. What a model looks at

      Features: the clues in the data a model uses to make its guess.

      3 min

   3. Training vs using

      A model learns once, then gets used many times. The world changes in between.

      3 min

   4. How AI learns

      Guess, see how wrong it was, adjust a little — repeated millions of times.

      3 min

   5. Memorising vs learning

      Overfitting: a model that aces the practice questions and fails the new ones.

      3 min

   6. More data helps, up to a point

      Why the quality and variety of data often matter more than the amount.

      3 min

   7. Garbage in, garbage out

      Wrong labels, missing data and lopsided examples all get learned too.

      3 min

   8. Recap: patterns, not magic

      The whole picture of how AI learns, in three minutes.

      3 min
4. Unit 4·6 bites·About 18 min

   Blind spots

   Where AI goes wrong — invention, staleness, bias — and how to check it.

   1. Confidently wrong

      Hallucination: when AI invents facts, quotes or sources that sound convincing.

      3 min

   2. Out of date

      A model's knowledge cutoff, and when you need a tool that can look things up.

      3 min

   3. Bias in, bias out

      Models inherit the imbalances in their data, including in decisions about people.

      3 min

   4. Sounds right is not the same as right

      Why fluent writing lowers our guard, and how to push back.

      3 min

   5. How to verify an answer

      Check the source, check the numbers, check with someone who knows — a simple routine for any answer that matters.

      3 min

   6. When not to use AI

      Tasks where the cost of a mistake outweighs the time saved.

      3 min
5. Unit 5·7 bites·About 21 min

   Data basics

   Reading numbers and charts honestly — the skill that makes you hard to fool, by people or by AI.

   1. What counts as data

      Numbers, text, images, clicks: data is a record of something that happened.

      3 min

   2. Rows and columns

      The table shape almost every analysis uses, and why one row should mean one thing.

      3 min

   3. Averages that mislead

      Mean vs median, and why "the average salary" often describes nobody.

      3 min

   4. Percent vs percentage points

      Going from 2% to 3% is up 1 point — or up 50%. Both are true, and that is the problem.

      3 min

   5. Correlation is not causation

      Two things rising together do not mean one causes the other.

      3 min

   6. Charts that mislead

      Truncated axes, cherry-picked ranges, and how to read a chart in ten seconds.

      3 min

   7. Sample size matters

      Why "3 out of 4 customers loved it" means something very different from "300 out of 400".

      3 min
6. Unit 6·6 bites·About 18 min

   Working with AI

   The everyday habits that separate people who use AI well from people who merely use it.

   1. Pick the right task

      Frequent, easy to check and cheap to get wrong: the best places to start.

      3 min

   2. AI drafts, you edit

      A clear division of labour: AI for speed, you for judgement.

      3 min

   3. Summaries you can trust

      How to ask for summaries that point back to the source, and how to check them.

      3 min

   4. Analysing a spreadsheet with AI

      Asking your data questions in plain language — and checking the numbers that come back.

      3 min

   5. Keeping a human in the loop

      Where a decision has to stay with a person, and why.

      3 min

   6. Your AI habit

      A small weekly routine that keeps the skill growing.

      3 min

Read while you wait

Guides from the Datamew team, free on the web today.

AI fundamentals

How to Write a Good Prompt: A 5-Part Framework for Work

A good prompt gives the AI five things: context, task, input, format and limits. It says who the work is for, what to do, which material to use, how the answer should look and what to avoid. Then you check the answer and ask a follow-up.

7 min read

AI from zero

How to Learn AI From Zero, Without Coding

Yes, you can learn AI without writing any code. To use AI well at work you need four things: a basic idea of how it works, a sense of where it fails, practice with one or two free tools, and the habit of checking every answer before you use it.

8 min read