Bi-directionality in Human-AI Collaborative Systems

The Symposium addresses the challenges in creating synergistic human and AI-based autonomous system-of-systems. Recent advances in generative AI techniques, such as Large Language Models, have exacerbated the growing concerns associated with AI, such as the risk, trust, and safety from the use of machines/AI in open situations. These concerns present major hurdles in the development of verified and validated engineered systems involving bi-directional pathways across the human-machine barrier; bi-directionality in this context means understanding the design and operational consequences of the human on the agent, and vice-versa. Current discussions on human-AI interactions are fragmented, focusing either on the impact of AI on human stakeholders (and relevant human factors considerations), or potential ways of involving humans in computational interventions (e.g., data annotation, behavior interpretation). We believe the challenges associated with humans-AI collaborative systems cannot be adequately addressed if the underlying challenges associated with bi-directionality are not taken into consideration.