Artificial Agents, Cloud Computing, and Quantum Computing: Applying Floridi’s Method of Levels of Abstraction

In his paper “On the Intrinsic Value of Information Objects and the Infosphere,” Luciano Floridi asserts that the goal of Information Ethics (IE) “is to fill an ‘ethical vacuum’ brought to light by the ICT revolution, to paraphrase Moor” (1985). He claims “IE will prove its value only if its applications bear fruit. This is the work that needs to be done in the near future” (Floridi 2002). Our chapter proposes to do part of that work. Initially we focus on Floridi’s Method of Levels of Abstraction (LoA). We begin by examining his methodology as it was first developed with J. W. Sanders in “The Method of Abstraction” (Floridi and Sanders 2004) and expanded in “The Method of Levels of Abstraction” (Floridi 2008a, b). Then we will demonstrate the general applicability and ethical utility of the method of levels of abstraction by considering three different computational paradigms: artificial agents, cloud computing, and quantum computing. In particular, we examine artificial agents as systems that embody the traditional digital computer (modeled as a single Turing machine). This builds on previous work by Floridi and Sanders (2004) and Grodzinsky et al. (2008). New contributions of this chapter include the application of the method of levels of abstraction to the developing paradigm of cloud computing and to the nascent paradigm of quantum computing. In all three paradigms, we emphasize aspects that highlight ethical issues.