AI Glossary
Risk Mitigation
The process of reducing the likelihood or impact of identified risks.
Risk Mitigation
Overview
Identifying a risk is only the first step.
Organizations must also decide what actions to take to reduce that risk.
This process is known as risk mitigation.
Risk mitigation refers to the strategies, controls, and actions implemented to reduce either the likelihood that a risk will occur or the impact if it does occur.
A helpful way to think about risk mitigation is installing smoke detectors in a building.
Smoke detectors do not prevent fires from starting.
They help detect problems early and reduce the potential consequences.
Organizations apply the same principle when managing AI systems.
Risk mitigation may include governance policies, security controls, human oversight, testing, model monitoring, employee training, or access restrictions.
Different risks require different mitigation strategies.
The goal is not eliminating every possible risk.
Instead, organizations seek to reduce risks to a level they consider acceptable while continuing to benefit from AI.
Why It Matters
Risk mitigation helps organizations reduce the impact of AI-related risks before they become significant problems.
Real-World Example
An organization requires human review for AI-generated hiring recommendations to reduce the risk of biased decisions.