Value Engineering in AI (VALE 2023)

With AI’s impact on our everyday lives becoming more and more tangible, the need for ethical and trustworthy AI has gained strong recognition by governments, industry, the general public, as well as academics, as evidenced by the numerous ethical guidelines and requirements. The stated goal of the EU strategy is to develop AI in such a way that applications using AI are reliable, robust, explainable, ethically guided and hence trustworthy. Ensuring AI is trustworthy and reliable requires ensuring that it fulfils human needs and respects human values. To achieve this, there is a need to develop software systems that reason about human values and norms, implement these values through norms, and ensure the alignment of behaviour with those values and norms. We argue that just as values guide our own morality, values can guide the morality of software agents and systems, bringing machine morality closer to reality. The result would be value-aware systems that take value-aligned decisions, interpret human behaviour in terms of values and enrich human reasoning by enhancing the human's value-awareness. 

Today there is a growing wealth of work in the field of AI on accounting for human values and working towards value-aligned behaviour, such as how AI learns human values, how individual values can be aggregated to the level of groups, how arguments that explicitly reference values can be made, how decision making can be value-driven, and how norms are selected to maximise value-alignment. VALE 2023 intends to bring in research on value engineering together and foster in-depth discussions on the topic.

Topic Areas: 

The relevant topics include, but are not limited to: 

• Value and norm representation 

• Value and norm learning 

• Value and norm agreement 

• Value and norm conflict resolution 

• Value-driven argumentation and negotiation 

• Value-driven decision making 

• Value-driven system design 

• Value-alignment 

• Value-driven explainability 

• Legal questions in value and norm enforcement