Introduction

Our values shape our behavior and understanding of ourselves and others. Just as values are fundamental to human interaction, imbuing AI systems with an understanding of values is crucial for building trustworthy and ethical artificial intelligence. What we consider self-evident in human interactions must be explicitly programmed into AI.

The VALAWAI project addresses this critical need by providing developers with a comprehensive toolbox. This toolbox empowers them to infuse AI applications with mechanisms that promote value-awareness and ensure alignment with human values, ultimately leading to more responsible and trustworthy AI.

This documentation provides information about the VALAWAI toolbox and its components. The documentation is organized into the following sections:

• Architecture: Describes the VALAWAI architecture based on the GNW neuron model.
• Use Cases: Describes the use cases that have guided the VALAWAI project's scientific research and from which the components of the VALAWAI toolbox have been developed.
• Components: Provides information about how components are defined and details the available components that make up the toolbox.
• Tutorials: Provides instructions on how to create a component or use the VALAWAI architecture.
• Demonstrations: Provides scripts demonstrating how to use the VALAWAI toolbox to run value-aware applications.
• Glossary: Defines terms and concepts used in the documentation.