What Matters: Matthias Scheutz on Ethical Guardrails for AI

Hear from our faculty, in the comfort of your own home or office, about what inspires them, and what they find most exciting in their research!

Artificial intelligence technology is increasingly penetrating every aspect of human societies and it is critical to ensure that AI-driven systems are aware and behave according to our social and moral norms. Unfortunately, current AI systems either are completely oblivious or they are really bad at recognizing social and moral expectations and living up to them. Professor Matthias Scheutz will provide an overview of the efforts of the Human-Robot Interaction Laboratory at Tufts to develop ethical AI-based systems that have explicit norm representations and are able to reason with them to make the decisions they can justify to humans in natural language dialogues. He will illustrate their work with various demonstrations of robots interacting with humans in natural language.