Human Centered AI in Digitized Working Environments
Research questions addressed in this project are: How can user acceptance of AI be fostered in companies and what are the implications for AI systems? Which factors are specifically relevant in different work contexts? What are relevant fields of AI application and what is the position of humans and AI systems within an organization and its network? What influences human-system interaction and how can individuals’ expectations be shaped to foster AI acceptance?
The rapid scientific and technological progress in the field of AI threatens to widen this gap while people fear for their jobs. In this context we focus on working environments which require in-depth domain knowledge and working experience provided by humans so that data-driven AI systems serve as assistive decision support systems. As example scenarios we consider AI-enhanced systems in different working domains, e.g., smart maintenance in industrial processes, chatbot based customer service and AI-aided software engineering.
The aim of this project is to work out a roadmap for a follow-up flagship project for tackling the acceptance and awareness gap of artificial intelligence (AI) in digitized working environments.
We tackle this problem by an interdisciplinary approach that considers all essential aspects at the human-AI-system interface in such working environments. In particular, we take into account a) psychological aspects regarding motivation to cooperate, b) transparency and trustworthiness of the AI system, and c) utilization of human experiential knowledge for AI systems by means of knowledge-based technologies on the one hand and more generalizing machine learning techniques on the other hand. As core innovation we aim at designing a human-AI teaming framework that allows the AI system to improve its learning and generalization as well as its explanatory capabilities.
As result we will obtain a more systematic overview of industrial areas where the effective interplay between process, AI component and human actors requires the conception of an evolutionary and continuous process in contrast to a complete automation without additional human interaction after a first temporary setup phase. In this context, we will obtain valuable insight into the as-is situation in the field and the optimization potential to adopt AI initiatives effectively.
This interdisciplinary project will be aligned with national and European AI initiatives such as the Digitial Innvoation Hub for AI and combines expertise from the fields of AI, in particular machine learning and knowledge representation, predictive analytics and maintenance, work psychology, machine perception, human centered computing and production.
Project Key Information
- Start of project : October1st, 2019, duration: 12 Monate
- Consortium: 5 partners
- Project budget: 256k Euro, thereof 200k Euro project funding
The project "AI@Work läuft" is funded in the frame of the program "IKT der Zukunft - 7. Ausschreibung 2018" of the Austrian Research Promotion Agency (FFG) and is cofunded by the Federal Ministry for Transport, Innovation and Technology.
Partners in the flagship project "Connecting Austria" are:
- Inst. für Arbeitsforschung und Arbeitspolitik, JKU
- Profactor GmbH
- apollo.ai GmbH
- Vienna University of Economics and Business
- Plattform Industrie 4.0