AI Safety & Responsible AI
Bias
A systematic tendency that can cause unfair or distorted outputs for certain people or situations.
Example
A résumé model favors wording more common in one demographic group's historical applications.
Why it matters
It helps people name and examine patterns that distort decisions or disadvantage users.
You'll hear it
“The errors fall much more often on this group.”
Can you control it?
Not directly
No, not directly. Bias describes a learned property, observed result, or broader concept rather than a normal user setting. You may influence it indirectly by changing the tool, data, prompt, or workflow.
Common questions
- Does removing a sensitive field remove all bias?
- No. Other information can indirectly reflect the same characteristic.
- Must bias be intentional?
- No. Historical patterns and design choices can create uneven effects without deliberate discrimination.
- Can there be more than one source of bias?
- Yes. Collection, labeling, design, use and interpretation can all contribute.