Guides

Short explainers built from the same evidence as the rest of the site. Facts and our interpretation are kept apart.

• Do you really need to pay for AI?

  The honest answer depends on your task and on evidence we can check. Here is what the tests do and do not tell you.

• Why the most expensive AI isn't always the best.

  Different tasks have different leaders. A model can top one test and sit mid-table on another.

• What AI benchmark scores actually mean.

  Two kinds of score, what each can tell you, and how to read a gap between two models.

• How to choose an AI for your small business.

  Emails, proposals and reports: what to check before you pay, and how not to confuse API charges with subscriptions.

• Which AI should students consider?

  Maths, writing and research have different evidence. What it supports and where to double-check.

• What makes one AI model different from another?

  Maker, version, size of the family, settings and access all change what you get.

• Why a model can score highly and still get things wrong.

  High scores are averages over a test. Your question is one question.

• How AI pricing works: tokens, “per million” and subscriptions.

  What a token is, why prices are quoted per million, why input and output cost different amounts, and why a chat subscription is a different thing.

• Models, apps and harnesses: what are you actually using?

  The AI model, the app you chat in and the tool wrapped around it are different things. What open weights means, and what our data can and cannot say about free access.

• What AI bias means, and what these tests can tell you.

  Bias can mean several different things. Which of them our data can show, which it cannot, and how we try to keep our own selection fair.