AI Ethics—A Bird’s Eye View
Open accessThe DOI matches its Crossref record; 45 of the 45 references it cites were found in a registry or an index.
Abstract
The explosion of data-driven applications using Artificial Intelligence (AI) in recent years has given rise to a variety of ethical issues regarding data collection, annotation, and processing using mostly opaque algorithms, as well as the interpretation and employment of the results of the AI pipeline. The ubiquity of AI applications negatively impacts a variety of sensitive areas, ranging from discrimination against vulnerable populations to privacy invasion and the environmental cost that these algorithms entail, and puts into focus on the ever present domain of AI ethics. In this review article we present a bird’s eye view approach of the AI ethics landscape, starting from a historical point of view, examining the moral issues that were introduced by big datasets and the application of non-symbolic AI algorithms, the normative approaches (principles and guidelines) to these issues and the ensuing criticism, as well as the actualization of these principles within the proposed frameworks. Subsequently, we focus on the concept of responsibility, both as personal responsibility of the AI practitioners and sustainability, meaning the promotion of beneficence for both the society and the domain, and the role of professional certification and education in averting unethical choices. Finally, we conclude with indicating the multidisciplinary nature of AI ethics and suggesting future challenges.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
High-performance medicine: the convergence of human and artificial intelligence
Topol, E. J. (2019) · Nature Medicine · doi:10.1038/s41591-018-0300-7 · Cited by 10,450 · cited 2×
“The most prominent subjects are lethal autonomous weapons systems [27], autonomous vehicles [28], and healthcare AI applications [29] (the last two areas often compared to each other as to the levels of autonomy [30], as well as the fostering of trust [31]).”
BackgroundDissecting racial bias in an algorithm used to manage the health of populations
Obermeyer, Z., Powers, B., Vogeli, C., & Mullainathan, S. (2019) · Science · doi:10.1126/science.aax2342 · Cited by 7,137 · cited 1×
“They also receive less care than they should, because patients with higher medical bills (typically white) are prioritized, since the algorithm deployed uses healthcare cost as a proxy for health, assuming that higher cost means more severe health problems [50].”
BackgroundOn the Dangers of Stochastic Parrots: Can Language Models Be Too Big? 🦜
Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021) · Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency · doi:10.1145/3442188.3445922 · Cited by 6,988 · cited 2×
“, in ad targeting [45] and natural language processing (NLP) [46], geo-diversity issues [47], and abusive NLP [48], among others (for a non-academic presentation of impacted sectors see [1], from which the title of this subsection is borrowed).”
BackgroundThe global landscape of AI ethics guidelines
Jobin, A., Ienca, M., & Vayena, E. (2019) · Nature Machine Intelligence · doi:10.1038/s42256-019-0088-2 · Cited by 6,220
Floridi, L., Cowls, J., Beltrametti, M., Chatila, R., Chazerand, P., Dignum, V., Luetge, C., Madelin, R., Pagallo, U., Rossi, F., et al. (2018) · Minds and Machines · doi:10.1007/s11023-018-9482-5 · Cited by 4,774
The ethics of algorithms: Mapping the debate
Mittelstadt, B. D., Allo, P., Taddeo, M., Wachter, S., & Floridi, L. (2016) · Big Data & Society · doi:10.1177/2053951716679679 · Cited by 4,567 · cited 1×
“This framework is mirrored partly to the types of ethical concerns that were raised by algorithms that were identified by Mittelstadt et al. [43]: three epistemic (having to do with the data and model deployed), namely, inconclusive, inscrutable or misguided evidence; and two normative ones, i.e.,…”
MethodThe Struggle to Govern the Commons
Dietz, T., Ostrom, E., & Stern, P. C. (2003) · Science · doi:10.1126/science.1091015 · Cited by 4,209 · cited 1×
“However, data ecosystems present difficulties with what is considered effective commons governance/ Specifically, according to Dietz [94], as cited by Taylor and Purtova in [91] (p. 3) “effective commons governance is easier to achieve when the following conditions are present: (i) the resource and…”
MethodEthical Decision Making by Individuals in Organizations: An Issue-Contingent Model
Jones, T. M. (1991) · The Academy of Management Review · doi:10.2307/258867 · Cited by 4,135 · cited 1×
“The authors identify the characteristics of a “bad case” by adopting the concept of moral intensity that was proposed by Jones [101] (p. 372) as “a construct that captures of the extent of issue-related moral imperative in a situation”.”
BackgroundBarocas, S., & Selbst, A. D. (2014) · doi:10.2139/ssrn.2477899 · Cited by 3,048 · cited 1×
“The advent and subsequent flourish of data science and Big Data technologies, especially when coupled with AI, focused the conversation on data, their production, manipulation, sharing, safekeeping, etc. [33–36].”
BackgroundGebru, T., Morgenstern, J., Vecchione, B., Vaughan, J. W., Wallach, H., Daumé, H., & Crawford, K. (2018) · arXiv · doi:10.48550/arxiv.1803.09010 · Cited by 2,474
Cited by
Teaching AI Ethics in Medical Education: A Scoping Review of Current Literature and Practices
Weidener, L., & Fischer, M. (2023) · Perspectives on Medical Education · doi:10.5334/pme.954 · Cited by 133
“The inconsistency in the scientific definition of AI is also reflected in the attempt to define AI ethics [29].”
BackgroundProposing a Principle-Based Approach for Teaching AI Ethics in Medical Education
Weidener, L., & Fischer, M. (2024) · JMIR Medical Education · doi:10.2196/55368 · Cited by 90
Ortega-Bolaños, R., Bernal-Salcedo, J., Germán Ortiz, M., Galeano Sarmiento, J., Ruz, G. A., & Tabares-Soto, R. (2024) · Artificial Intelligence Review · doi:10.1007/s10462-024-10740-3 · Cited by 76
AI Ethics in the Fields of Education and Research: A Systematic Literature Review
Contreras, M. R., & Jaimes, J. O. P. (2024) · 2024 International Symposium on Accreditation of Engineering and Computing Education (ICACIT) · doi:10.1109/icacit62963.2024.10788651 · Cited by 18
Metadata
- Authors
- Christoforaki, M., & Beyan, O.
- Venue
- Applied Sciences
- Publisher
- MDPI AG
- Published
- 2022-04-20
- Volume
- 12
- Issue
- 9
- Pages
- 4130
- ISSN
- 2076-3417 (ISSN-L)
- OpenAlex
- W4224256546
- Cited by
- 50
- References
- 45
- FWCI
- 10.80
- Type
- journal article
- Language
- en
- Open access
- gold
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.