Bias occurs when an AI system produces unfair, skewed, or discriminatory results due to biased training data, flawed design, or human input.
Why it matters:
- Can lead to real-world harm (e.g., biased hiring algorithms or loan approvals).
- Erodes trust in AI systems.
- A major ethical and regulatory concern.
Pro tip: Always ask vendors how they test for and mitigate bias. Diverse, well-curated training data is the first line of defense.
« Back to Glossary Index