WAICOM – Workshop on AI Compliance Mechanism (full day)
With the development and spread of AI techniques, ensuring the adherence of AI’s behavior to legal and ethical principles has become a major subject. General fear of the unintended effects of AI systems, by its actions and its use of personal data, has led to a strong demand for trustworthy AI. This is a central concern that has become prominent both in public opinion and policy maker’s agenda. The EU High-Level Expert Group on AI, recently convened by the European Commission, published a report on “Ethics Guidelines for Trustworthy AI” says that AI systems should be: – lawful, complying with all applicable laws and regulations – ethical, ensuring adherence to ethical principles and values. The purpose of this workshop is to bring researchers together to present approaches to tackling legal/ethical AI compliance problems and to discuss selected challenges arising from AI compliance. We also solicit use cases related with AI compliance problems to create a basis to investigate common problems for future collaboration.
Programme
09:00-09:05
Opening Remark
09:05-09:30
Driving-Decision Making of Autonomous Vehicle according to Queensland Overtaking Traffic Rules (Hanif Bhuiyan, Guido Governatori and Andry Rakotonirainy), long
09:30-09:55
An Ontology-based Model for Handling Rule Exceptions in Traffic Scenes (Ya Wang, Maximilian Grabowski and Adrian Paschke), long
09:55-10:20
Ontology to Represent Norms : Ethics, Law and Informatics (Anaelle Martin), long
10:20-10:45
Devising Regulatory Sandboxes and Responsible Practices for Designing AI-based Services in the Finnish Public Sector (Nitin Sawhney and Ana Paula Gonzalez Torres), long
10:45-11:00 Coffee break
11:00-11:25
Dealing with differentiated modalities of the Good: beyond Pareto optimality (Guillaume Gervois, Gauvain Bourgne and Marie-Jeanne Lesot), long
11:25-11:50
Normative Diagrams (Diogo Sasdelli), long
11:50-12:15
Debugging Constraint Hierarchies Representing Ethical Norms with Valuation Preferences (Wachara Fungwacharakorn, Kanae Tsushima and Ken Satoh), long
12:15-12:40
Towards Continuous Audit-based Certification for MLOps (Dorian Knoblauch and Jürgen Großmann), long
12:40-14:10 Lunch
14:10-14:30
The Rigorous Approach to Process Compliance (Guido Governatori), published
14:30-14:50
Modeling a GDPR Compliant Data Wallet Application in Prova and AspectOWL (Ralph Schäfermeier, Theodoros Mitsikas and Adrian Paschke), published
14:50-15:10
Compliance through Model Checking (Avishkar Mahajan, Martin Strecker, Joe Watt and Meng Weng Wong), short
15:10-15:30
Obligation for AI systems in Healthcare: prepare for trouble and make it double? (Marinella Quaranta, Ilaria Angela Amantea and Marco Grosso), short
15:30-15:45 Cofeee break
15:45-16:05
Action Languages Based Actual Causality in Decision Making Contexts (Camilo Sarmiento, Gauvain Bourgne, Katsumi Inoue, and Jean-Gabriel Ganascia), published
16:05-16:25
Modeling Administrative Discretion Using Goal-Directed Answer Set Programming (Joaquín Arias), published
16:25-16:45
AIRO: An Ontology for Representing AI Risks Based on the Proposed EU AI Act and ISO Risk Management Standards (Delaram Golpayegani, Harshvardhan J. Pandit and Dave Lewis), published
16:45-17:00 Closing Remark