TAAFT
Free mode
100% free
Freemium
Free Trial
Deals
Create tool
Glossary
Ethics & Safety
AI and Bias

AI and Bias

[eɪ.aɪ ænd ˈbaɪəs]
Ethics & Safety
Last updated: December 9, 2024

Definition

Systematic errors or unfair prejudices embedded in AI systems due to training data or algorithmic design.

Detailed Explanation

Details how bias can enter AI systems through training data, algorithm design, or deployment contexts. Includes types of bias (selection, confirmation, sampling), detection methods, and mitigation strategies.

Use Cases

Hiring systems, loan approval, criminal justice risk assessment

Related Terms

Ethical Auditing

Systematic assessment of AI systems for compliance with ethical principles and guidelines.

Ethics & Safety

Transparency

The openness and clarity about how AI systems work, make decisions, and use data.

Ethics & Safety

Data Protection

Safeguarding of personal and sensitive information used in AI systems through technical and organizational measures.

Ethics & Safety