Better, Nicer, Clearer, Fairer: A Critical Assessment of the Movement for Ethical Artificial Intelligence and Machine Learning

By Daniel Greene, Anna Lauren Hoffmann, Luke Stark. 2019

324 citations recorded by the source corpus.

This paper uses frame analysis to examine recent high-profile values statements endorsing ethical design for artificial intelligence and machine learning (AI/ML). Guided by insights from values in design and the sociology of business ethics, we uncover the grounding assumptions and terms of debate that make some conversations about ethical design possible while forestalling alternative visions. Vision statements for ethical AI/ML co-opt the language of some critics, folding them into a limited, technologically deterministic, expert-driven view of what ethical AI/ML means and how it might work.

Subfields: Cognitive Neuroscience, Safety Research

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

In Safety Research

Citations in the corpus

Closing the AI accountability gap: defining an end-to-end framework for internal algorithmic auditing (2020, positive): “little to no means of accountability [27, 82].”
Decolonial AI: Decolonial Theory as Sociotechnical Foresight in Artificial Intelligence (2020, background): “Scholarship in this domain (Winner, 1980; Nissenbaum, 2001; Greene et al., 2019) aims not only to explain sociotechnical phenomena, but to also examine issues of values, culture and power at play between stakeholders and technological artefacts.”
Data Governance in the Age of Large-Scale Data-Driven Language Technology (2022, background): “As such, devising a governance structure based solely on values expressed by our participants runs the risk of prioritizing their interests and excluding visions that may be more relevant to other language users [65].”
ABOUT ML: Annotation and Benchmarking on Understanding and Transparency of Machine Learning Lifecycles (2019, background): “Although study after study confirms that meaningful progress cannot be made until ethical ideals are operationalized [17, 12, 5], the inconsistency with which high level principles such as transparency are interpreted across different contexts, organizations and even teams makes it difficult to desi”
Algorithmic realism: expanding the boundaries of algorithmic thought (2020, background): “reasoning, e orts to promote ethics within computer science and the tech industry have tended to follow a narrow logic of technological determinism and technological solutionism [44, 59, 96, 150].”
The Ethics of AI Ethics: An Evaluation of Guidelines (2019, background): “2018) over meta-studies about AI ethics (Vakkuri and Abrahamsson 2018; Prates et al. 2018; Boddington 2017; Greene et al. 2019; Goldsmith and Burton 2017) or the empirical analysis on how trolley problems are solved (Awad et al.”
Terms of inclusion: Data, discourse, violence (2020, background): “As Greene, Hoffmann, and Stark (2019) have shown in their work on values and ethics statements guiding the development of machine learning and AI, these efforts provide little indication that data technologies can be limited or constrained—they paint ethics as a matter of iterative improvement and…”
Artificial Intelligence Regulation: A Meta-Framework for Formulation and Governance (2020, unknown): “When analyzing several movements advocating the establishment of criteria for how to best use AI, studies identified an opportunity to develop a competition around a technological reform [50].”
Confronting Power and Corporate Capture at the FAccT Conference (2022, background): “Indeed, Big Tech diagnoses problems as purely technical, rather than a source for policy action [28]; Ganesh and Moss find that harms identified by marginalized groups are predominately subsumed into design fixes [23].”
Limits and Possibilities for “Ethical AI” in Open Source: A Study of Deepfakes (2022, background): “Systems may adhere to such principles while still being patently unethical [42], and convergence on principles risks obscuring political and normative disagreements [55], or focuses “Ethical AI” scrutiny on AI design rather than the business uses it enables [30].”
A Framework for Deprecating Datasets: Standardizing Documentation, Identification, and Communication (2021, background): “Such research describes complex ethical, social, and political considerations with which the ML community and field at large are beginning to grapple [36, 44, 94].”