Toolkit for Specification, Validation and Verification of Social, Legal, Ethical, Empathetic and Cultural Requirements for Autonomous Agents

Sinem Getir Yaman*, Pedro Ribeiro, Charlie Burholt, Maddie Jones, Ana Lucia Caneca Cavalcanti, Radu Calinescu

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

A growing range of applications use AI and other autonomous agents to perform tasks that raise social, legal, ethical, empathetic, and cultural (SLEEC) concerns. To support a framework for the consideration of these concerns, we introduce SLEEC-TK, a toolkit for specification, validation, and verification of SLEEC requirements. SLEEC-TK is an Eclipse-based environment for defining SLEEC rules in a domain-specific language with a timed process algebraic semantics. SLEEC-TK uses model checking to identify redundant and conflicting rules, and to verify conformance of design models with SLEEC rules. We illustrate the use of SLEEC-TK for an assistive-care robot.
Original languageEnglish
Pages (from-to)103118
Number of pages6
JournalScience of Computer Programming
Volume236
Early online date16 Apr 2024
DOIs
Publication statusE-pub ahead of print - 16 Apr 2024

Bibliographical note

Accepted as an Original Software Publication (OSP), under the Software Track of Science of Computer Programming.

© 2024 The Authors

Keywords

  • autonomous agents
  • social requirements
  • legal requirements
  • ethical requirements
  • empathetic requirements
  • cultural requirements
  • verification
  • validation

Cite this