AI Principles

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

Curiosity-driven learning emphasizes exploration and intrinsic motivation in the learning process, fostering deeper understanding.

Deliberative Alignment ensures AI systems reflect human values through collaborative decision-making processes.

Empowerment refers to the process of enabling individuals or groups to gain control over their lives and make informed decisions.

Group Fairness ensures that AI systems treat different demographic groups equitably.

Individual fairness ensures similar individuals receive similar treatment in AI systems.

Intrinsic motivation refers to engaging in activities for their own sake, driven by internal rewards rather than external pressures.

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

Marginalization refers to the process by which certain groups or individuals are pushed to the edge of society, limiting their access to resources.

Moral Uncertainty Modelling addresses decision-making under conflicting moral values using AI techniques.

Occam's Razor is a problem-solving principle stating that the simplest explanation is usually the best.

A mindset that embraces positivity despite uncertain outcomes, especially in decision-making and problem-solving contexts.

Parameter Transparency refers to the clarity and accessibility of AI model parameters to users and developers.

Pareto efficiency is an economic concept where resources are allocated in a way that no one can be made better off without making someone else worse off.

A Pareto Improvement occurs when a change benefits at least one individual without making anyone worse off.

The Pareto Principle states that 80% of effects come from 20% of causes, often applied in business and economics.