AI Governance

Accountability is the obligation to explain, justify, and take responsibility for actions and decisions, particularly in AI systems.

A framework categorizing AI systems based on their alignment with human values and intentions.

Constitutional Prompting is a method for ensuring AI behavior aligns with human values and ethical guidelines.

Data Governance is a framework for managing data availability, usability, integrity, and security within organizations.

A Default Policy is a preset rule used by AI systems to handle situations not explicitly defined in their programming.

Demographic parity ensures equal outcomes across different demographic groups in AI decision-making.

Dual-Use Risk refers to the potential for technologies to be used for both beneficial and harmful purposes.

Fairness in AI refers to the impartial treatment of individuals or groups in algorithmic decision-making.

Human Oversight refers to the involvement of people in monitoring and guiding AI systems to ensure ethical and accurate decision-making.

Machine Ethics is the study of moral principles guiding AI behavior and decision-making.

A Model Register is a centralized database for managing AI models throughout their lifecycle.

Model Risk Management involves identifying, assessing, and mitigating risks associated with predictive models in AI applications.

Scalable Oversight refers to systems that can effectively manage and monitor AI as it grows in complexity and usage.

A scripted policy is a predefined set of automated rules guiding system behavior in AI applications.