Skip to content
Vol. 1 · Weekly editionWeekly · in your inbox
inklaretaal
Friday, 4 September 2026 · Clarity since 2026
Glossary

What is Bias?

Bias is a skewed or one-sided picture an AI has picked up from its examples, which can make it respond unfairly or with prejudice.

A model learns from examples, and those examples are full of human patterns. If something shows up in the data more often in a certain way, the model takes that on. So an AI might picture a man more readily for 'director', simply because that is how it appeared in the data. That is bias.

What this means for you

For decisions about people (hiring, reviewing, selecting), bias can do real harm. Use AI there as a help, never as the final verdict, and keep looking critically yourself.

Put precisely

Bias is not a bug in the code. The model picks up the patterns that were in the examples, and those examples came from people.

Common misconception

That you remove bias by banning a few words. The pattern sits in the numbers, not in a list of words.


Related words

Training dataHallucinationAI

Understand AI, calmly

This word is a start. The inklaretaal Learning curve explains AI step by step in plain language, tuned to your field. The first month is free.

Start free →Back to the glossary