Current advances in research, development and application of artificial intelligence systems have yielded a far-reaching discourse on AI ethics. In consequence, a number of ethics guidelines have been released in recent years. These guidelines comprise normative principles and recommendations aimed to harness the “disruptive” potentials of new AI technologies. Designed as a semi-systematic evaluation, this paper analyzes and compares 22 guidelines, highlighting overlaps but also omissions. As a result, I give a detailed overview of the field of AI ethics. Finally, (...) I also examine to what extent the respective ethical principles and values are implemented in the practice of research, development and application of AI systems—and how the effectiveness in the demands of AI ethics can be improved. (shrink)

The debate about the ethical implications of Artificial Intelligence dates from the 1960s :741–742, 1960; Wiener in Cybernetics: or control and communication in the animal and the machine, MIT Press, New York, 1961). However, in recent years symbolic AI has been complemented and sometimes replaced by Neural Networks and Machine Learning techniques. This has vastly increased its potential utility and impact on society, with the consequence that the ethical debate has gone mainstream. Such a debate has primarily focused on principles—the (...) ‘what’ of AI ethics —rather than on practices, the ‘how.’ Awareness of the potential issues is increasing at a fast rate, but the AI community’s ability to take action to mitigate the associated risks is still at its infancy. Our intention in presenting this research is to contribute to closing the gap between principles and practices by constructing a typology that may help practically-minded developers apply ethics at each stage of the Machine Learning development pipeline, and to signal to researchers where further work is needed. The focus is exclusively on Machine Learning, but it is hoped that the results of this research may be easily applicable to other branches of AI. The article outlines the research method for creating this typology, the initial findings, and provides a summary of future research needs. (shrink)

Frequent public uproar over forms of data science that rely on information about people demonstrates the challenges of defining and demonstrating trustworthy digital data research practices. This paper reviews problems of trustworthiness in what we term pervasive data research: scholarship that relies on the rich information generated about people through digital interaction. We highlight the entwined problems of participant unawareness of such research and the relationship of pervasive data research to corporate datafication and surveillance. We suggest a way forward by (...) drawing from the history of a different methodological approach in which researchers have struggled with trustworthy practice: ethnography. To grapple with the colonial legacy of their methods, ethnographers have developed analytic lenses and researcher practices that foreground relations of awareness and power. These lenses are inspiring but also challenging for pervasive data research, given the flattening of contexts inherent in digital data collection. We propose ways that pervasive data researchers can incorporate reflection on awareness and power within their research to support the development of trustworthy data science. (shrink)

In this paper, we discuss several problems with current Big data practices which, we claim, seriously erode the role of informed consent as it pertains to the use of personal information. To illustrate these problems, we consider how the notion of informed consent has been understood and operationalised in the ethical regulation of biomedical research (and medical practices, more broadly) and compare this with current Big data practices. We do so by first discussing three types of problems that can impede (...) informed consent with respect to Big data use. First, we discuss the transparency (or explanation) problem. Second, we discuss the re-repurposed data problem. Third, we discuss the meaningful alternatives problem. In the final section of the paper, we suggest some solutions to these problems. In particular, we propose that the use of personal data for commercial and administrative objectives could be subject to a ‘soft governance’ ethical regulation, akin to the way that all projects involving human participants (e.g., social science projects, human medical data and tissue use) are regulated in Australia through the Human Research Ethics Committees (HRECs). We also consider alternatives to the standard consent forms, and privacy policies, that could make use of some of the latest research focussed on the usability of pictorial legal contracts. (shrink)

BackgroundEthics review is the process of assessing the ethics of research involving humans. The Ethics Review Committee (ERC) is the key oversight mechanism designated to ensure ethics review. Whether or not this governance mechanism is still fit for purpose in the data-driven research context remains a debated issue among research ethics experts.Main textIn this article, we seek to address this issue in a twofold manner. First, we review the strengths and weaknesses of ERCs in ensuring ethical oversight. Second, we map (...) these strengths and weaknesses onto specific challenges raised by big data research. We distinguish two categories of potential weakness. The first category concerns persistent weaknesses, i.e., those which are not specific to big data research, but may be exacerbated by it. The second category concerns novel weaknesses, i.e., those which are created by and inherent to big data projects. Within this second category, we further distinguish between purview weaknesses related to the ERC’s scope (e.g., how big data projects may evade ERC review) and functional weaknesses, related to the ERC’s way of operating. Based on this analysis, we propose reforms aimed at improving the oversight capacity of ERCs in the era of big data science.ConclusionsWe believe the oversight mechanism could benefit from these reforms because they will help to overcome data-intensive research challenges and consequently benefit research at large. (shrink)

This paper sheds light on the role of digital platform labour in the development of today’s artificial intelligence, predicated on data-intensive machine learning algorithms. Focus is on the specific ways in which outsourcing of data tasks to myriad ‘micro-workers’, recruited and managed through specialized platforms, powers virtual assistants, self-driving vehicles and connected objects. Using qualitative data from multiple sources, we show that micro-work performs a variety of functions, between three poles that we label, respectively, ‘artificial intelligence preparation’, ‘artificial intelligence verification’ (...) and ‘artificial intelligence impersonation’. Because of the wide scope of application of micro-work, it is a structural component of contemporary artificial intelligence production processes – not an ephemeral form of support that may vanish once the technology reaches maturity stage. Through the lens of micro-work, we prefigure the policy implications of a future in which data technologies do not replace human workforce but imply its marginalization and precariousness. (shrink)

Ethics has powerful teeth, but these are barely being used in the ethics of AI today – it is no wonder the ethics of AI is then blamed for having no teeth. This article argues that ‘ethics’ in the current AI ethics field is largely ineffective, trapped in an ‘ethical principles’ approach and as such particularly prone to manipulation, especially by industry actors. Using ethics as a substitute for law risks its abuse and misuse. This significantly limits what ethics can (...) achieve and is a great loss to the AI field and its impacts on individuals and society. This article discusses these risks and then highlights the teeth of ethics and the essential value they can – and should – bring to AI ethics now. (shrink)

There are growing discontinuities between the research practices of data science and established tools of research ethics regulation. Some of the core commitments of existing research ethics regulations, such as the distinction between research and practice, cannot be cleanly exported from biomedical research to data science research. Such discontinuities have led some data science practitioners and researchers to move toward rejecting ethics regulations outright. These shifts occur at the same time as a proposal for major revisions to the Common Rule—the (...) primary regulation governing human-subjects research in the USA—is under consideration for the first time in decades. We contextualize these revisions in long-running complaints about regulation of social science research and argue data science should be understood as continuous with social sciences in this regard. The proposed regulations are more flexible and scalable to the methods of non-biomedical research, yet problematically largely exclude data science methods from human-subjects regulation, particularly uses of public datasets. The ethical frameworks for Big Data research are highly contested and in flux, and the potential harms of data science research are unpredictable. We examine several contentious cases of research harms in data science, including the 2014 Facebook emotional contagion study and the 2016 use of geographical data techniques to identify the pseudonymous artist Banksy. To address disputes about application of human-subjects research ethics in data science, critical data studies should offer a historically nuanced theory of “data subjectivity” responsive to the epistemic methods, harms and benefits of data science and commerce. (shrink)