What is Bias in AI

Definition

Bias in AI is a systematic error that causes a model to perform worse for some users, data, or scenarios. Simply put, this concept helps assess risk, liability, safety, and compliance. In practice, it helps to understand what capabilities the tool actually has, what data it will need, and what limitations are worth checking before implementation.

Example

A resume selection model may be more likely to underestimate candidates from a group that is poorly represented in the training data.

Why it matters

Bias directly impacts the trust, security, and legal risks of an AI product. This helps you choose AI tools not by big promises, but by how they work in a real problem.

How it works

First, stakeholders, data, and potential harm are identified, then checks, restrictions, audits, and responsibilities are introduced. In the case of the term AI Bias, it is important to look at the data, quality criteria, and application conditions separately.

Where it is used

• Important in products where AI impacts people, personal data, security, legal risks or decision making.

Limitations

Risks change as laws, products and data change, so these pages require regular editorial review.