Machine invention systems: a (r)evolution of the invention process?

Abstract

Current developments in fields such as quantum physics, fine arts, robotics, cognitive sciences or defense and security indicate the emergence of creative systems capable of producing new and innovative solutions through combinations of machine learning algorithms. These systems, called machine invention systems, challenge the established invention paradigm in promising the automation of – at least parts of – the innovation process. This paper’s main contribution is twofold. Based on the identified state-of-the-art examples in the above mentioned fields, key components for machine invention systems and their relations are identified, creating a conceptual model as well as proposing a working definition for machine invention systems. The differences and delimitations to other concepts in the field of machine learning and artificial intelligence, such as machine discovery systems are discussed as well. Furthermore, the paper briefly addresses the social and societal implications and limitations that come with the adoption of the technology. Because of their revolutionizing potential, there are widespread implications to consider from ethical and moral implications to policymaking and societal changes, like changes in the job structure. The discussion part approaches some of these implications, as well as solutions to some of the proposed challenges. The paper concludes by discussing some of the systemic benefits that can be accessed through machine invention.

Building Resilient Governance Frameworks for Human-Robot Collaboration: Towards a More Interdisciplinary Understanding of Risk and Ethics in European Regulation

“…novelty, expressiveness, exposition time or degree of anthropomorphismand the values they uphold -e.g. efficiency, reliability, cost minimization or sustainability (Vasilescu & Filzmoser, 2021). These elements are selected for analysis because they conform the basis of each author's interpretation of the relationship between the social and technical aspects of HRC, which in turn explains the regulatory and ethical framework they implicitly or explicitly invite (van de Poel, 2020;van Berkel, 2022).…”

“…Equally as widespread in the sources reviewed are fears that robots will soon curtail the agency of workers, consumers and even by-standers. The perspective of large-scale deployment of robots in the workplace is commonly associated with negative psychological effects on workers (Vasilescu & Filzmoser, 2021) and fears of downward pressure on working conditions, a rise in inequality, and unemployment (idem).…”

The consolidation of the digital technological system

“…They conclude that there was a reorientation towards applied deep learning solutions from the 2010s onwards. More than that, they conclude that DL systems alters knowledge-producing circles such as science Vasilescu and Filzmoser (2021). conceptualize machine invention systems, which share a similar view on the invention in the method of invention (IMI) Bianchini et al (2020).…”