The DOI matches its Crossref record; 72 of the 72 references it cites were found in a registry or an index; none of the 135 sources its claims cite is checked for support yet.
AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry
Open accessAbstract
A new and unorthodox approach to deal with discriminatory bias in Artificial Intelligence is needed. As it is explored in detail, the current literature is a dichotomy with studies originating from the contrasting fields of study of either philosophy and sociology or data science and programming. It is suggested that there is a need instead for an integration of both academic approaches, and needs to be machine-centric rather than human-centric applied with a deep understanding of societal and individual prejudices. This article is a novel approach developed into a framework of action: a bias impact assessment to raise awareness of bias and why, a clear set of methodologies as shown in a table comparing with the four stages of pharmaceutical trials, and a summary flowchart. Finally, this study concludes the need for a transnational independent body with enough power to guarantee the implementation of those solutions.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
94 Not checked yet · 31 Own interpretation · 10 No basis given
None of the sources cited for this claim have been checked yet.
In the paper
“One example can be when the word woman was penalised, by being given a negative or a lower weight, on a CV selection process based on the data of an industry traditionally dominated by men like the tech industry [79, 82].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data
What the sources say
- Not checked yetcites [79] Shields, M.: Women's participation in Seattle's high-tech economy. https://smartech.gatech.edu/bitstream/handle/1853/53790/madelyn_shields_womens_participation_in_seattles_hightech_economy.pdf (2015). Accessed 15 Aug 2021
- Not checked yetcites [82] Swift, S.: Gender Disparities in the Tech Industry: The Effects of Gender and Stereotypicability on Perceived Environmental Fit. In: 2015 NCUR (2015)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
None of the sources cited for this claim have been checked yet.
In the paper
“Whilst useful in creating debate, these tend to present either no solutions at all or overly simplified single solutions [13, 15, 19, 29, 32, 35, 56, 60, 64, 89, 96, 104].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Introduction
What the sources say
- Not checked yetcites [13] Binns, R.: Fairness in machine learning: lessons from political philosophy. In Conference on Fairness, Accountability and Transparency, pp. 149–159. PMLR (2018)
- Not checked yetcites [15] Boddington, P.: Towards a Code of Ethics for Artificial Intelligence, pp. 27–37. Springer, Cham (2017) source
- Not checked yetcites [19] Bostrom, N., Yudkowsky, E.: The ethics of artificial intelligence. Camb. Handb. Artif. Intell. 1, 316–334 (2014) source
- Not checked yetcites [29] Chalmers, D.: The singularity: a philosophical analysis. In: Schneider, S. (ed.) Science Fiction and Philosophy: From Time Travel to Superintelligence, pp. 171–224. Wiley, UK (2009) source
- Not checked yetcites [32] Crawford, K.: The Atlas of AI. Yale University Press (2021) source
- Not checked yetcites [35] Dwork, C., Hardt, M., Pitassi, T., Reingold, O., Zemel, R.: Fairness through awareness. In Proceedings of the 3rd Innovations in Theoretical Computer Science Conference, pp. 214–226 (2012) source
- Not checked yetcites [56] Müller, V.C. (Summer 2021 Edition), Zalta, E.N.: (eds.) Ethics of artificial intelligence and robotics. The Stanford Encyclopedia of Philosophy. https://plato.stanford.edu/archives/sum2021/entries/ethics-ai/. Accessed 18 Mar 2021
- Not checked yetcites [60] Noble, S.U.: Algorithms of Oppression. New York University Press, New York (2018) source
- Not checked yetcites [64] O’Neil, C.: Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy. Penguin Books Limited, New York (2016)
- Not checked yetcites [89] Van Wynsberghe, A., Robbins, S.: Critiquing the reasons for making artificial moral agents. Sci. Eng. Ethics 25(3), 719–735 (2019) source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“The quality of the collected data will influence the quality of the algorithmic decisions.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The second reason, incomplete or unrepresentative data, is a consequence of the first.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“The decision process tends to rely on previous data, its outcomes and the initial weight given by the programmer.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
None of the sources cited for this claim have been checked yet.
In the paper
“If the data are biased with one example being prejudicial bias—racial bias being a well-documented case [2, 23], the outcome will likely follow suit unless appropriate controls are put in place.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data
What the sources say
- Not checked yetcites [2] Angwin, J., et al.: (2016) Machine bias: there’s software used across the country to predict future criminals. And it’s biased against blacks. ProPublica. https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing (2016). Accessed 28 Mar 2021
- Not checked yetcites [23] Buolamwini, J., Gebru, T.: Gender shades: intersectional accuracy disparities in commercial gender classification. In: Conference on fairness, accountability and transparency, pp. 77–91. PMLR (2018)
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“The computer can be used as a proxy for a human, relinquishing them of any moral responsibility [60].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data
What the sources say
- Not checked yetcites [60] Noble, S.U.: Algorithms of Oppression. New York University Press, New York (2018) source
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“The training data are non-existent or incomplete.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data
What the sources say
- Not checked yetcites [84] Thelisson, E., Padh, K., Celis, L.E.: Regulatory mechanisms and algorithms towards trust in AI/ML. In: Proceedings of the IJCAI 2017 Workshop on Explainable Artificial Intelligence (XAI), Melbourne, Australia (2017)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“Part of the challenge of understanding algorithmic oppression is to understand that mathematical formulations to drive automated decisions are made by human beings”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data
What the sources say
- Not checked yetcites [60, p. 14]
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Even in an unsupervised ML model working with raw data, the machine might find discriminatory societal patterns and replicate them.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
None of the sources cited for this claim have been checked yet.
In the paper
“These benchmarks can be used in the evaluation of facial recognition systems that were biased towards skin colour and gender [23, 60].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Definition of artificial intelligence, machine learning, algorithms and AI bias
What the sources say
- Not checked yetcites [23] Buolamwini, J., Gebru, T.: Gender shades: intersectional accuracy disparities in commercial gender classification. In: Conference on fairness, accountability and transparency, pp. 77–91. PMLR (2018)
- Not checked yetcites [60] Noble, S.U.: Algorithms of Oppression. New York University Press, New York (2018) source
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“The process of data mining,4 one of the ways algorithms collect and analyse data [88], can already be discriminatory as a start, because it decides which data are of value, which is of no value and its weight—how valuable it is.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data
What the sources say
- Not checked yetcites [88] Van Wel, L., Royakkers, L.: Ethical issues in web data mining. Ethics Inf. Technol. 6(2), 129–140 (2004) source
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“The outcome ended discriminating women in the selection process [33].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data
What the sources say
- Not checked yetcites [33] Dastin, J.: Amazon scraps secret AI recruiting tool that showed bias against women. Reuters. https://www.reuters.com/article/us-amazon-com-jobs-automation-insight/amazon-scraps-secret-ai-recruiting-tool-that-showed-bias-against-women-idUSKCN1MK08G (2018). Accessed 24 Apr 2021 source
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
None of the sources cited for this claim have been checked yet.
In the paper
“the current literature is a dichotomy with studies originating from the contrasting fields of study of either philosophy and sociology or data science and programming”
- Also stated in: § Introduction · plain
“Existing literature analysing AI bias tend to originate from one of two, very separate, academic spheres.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Abstract
What the sources say
- Not checked yetcites [13] Binns, R.: Fairness in machine learning: lessons from political philosophy. In Conference on Fairness, Accountability and Transparency, pp. 149–159. PMLR (2018)
- Not checked yetcites [15] Boddington, P.: Towards a Code of Ethics for Artificial Intelligence, pp. 27–37. Springer, Cham (2017) source
- Not checked yetcites [19] Bostrom, N., Yudkowsky, E.: The ethics of artificial intelligence. Camb. Handb. Artif. Intell. 1, 316–334 (2014) source
- Not checked yetcites [29] Chalmers, D.: The singularity: a philosophical analysis. In: Schneider, S. (ed.) Science Fiction and Philosophy: From Time Travel to Superintelligence, pp. 171–224. Wiley, UK (2009) source
- Not checked yetcites [32] Crawford, K.: The Atlas of AI. Yale University Press (2021) source
- Not checked yetcites [35] Dwork, C., Hardt, M., Pitassi, T., Reingold, O., Zemel, R.: Fairness through awareness. In Proceedings of the 3rd Innovations in Theoretical Computer Science Conference, pp. 214–226 (2012) source
- Not checked yetcites [56] Müller, V.C. (Summer 2021 Edition), Zalta, E.N.: (eds.) Ethics of artificial intelligence and robotics. The Stanford Encyclopedia of Philosophy. https://plato.stanford.edu/archives/sum2021/entries/ethics-ai/. Accessed 18 Mar 2021
- Not checked yetcites [60] Noble, S.U.: Algorithms of Oppression. New York University Press, New York (2018) source
- Not checked yetcites [64] O’Neil, C.: Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy. Penguin Books Limited, New York (2016)
- Not checked yetcites [89] Van Wynsberghe, A., Robbins, S.: Critiquing the reasons for making artificial moral agents. Sci. Eng. Ethics 25(3), 719–735 (2019) source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § Introduction · plain
In the paper
“Training data that are biased can lead to discriminatory ML models.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data
What the sources say
- Not checked yetcites [84] Thelisson, E., Padh, K., Celis, L.E.: Regulatory mechanisms and algorithms towards trust in AI/ML. In: Proceedings of the IJCAI 2017 Workshop on Explainable Artificial Intelligence (XAI), Melbourne, Australia (2017)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“This elite constantly develops strategies, either consciously or unconsciously, to prevent others from accessing their privileges [17]—and the elaboration of prejudices is one of them.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data
What the sources say
- Not checked yetcites [17] Bonilla-Silva, E.: White Supremacy and Racism in the Post-Civil Rights Era. Lynne Rienner Publishers, Boulder (2001) source
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: unclear — It is unclear whether it could be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
None of the sources cited for this claim have been checked yet.
In the paper
“In 2014, Amazon started to use an algorithm to select the top five CVs from one hundred applicants.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
What the sources say
- Not checked yetcites [33] Dastin, J.: Amazon scraps secret AI recruiting tool that showed bias against women. Reuters. https://www.reuters.com/article/us-amazon-com-jobs-automation-insight/amazon-scraps-secret-ai-recruiting-tool-that-showed-bias-against-women-idUSKCN1MK08G (2018). Accessed 24 Apr 2021 source
- Not checked yetcites [91] Vincent, J.: Amazon reportedly scraps internal AI recruiting tool that was biased against women. The Verge. https://www.theverge.com/2018/10/10/17958784/ai-recruiting-tool-bias-amazon-report (2018). Accessed 28 Mar 2021
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“The same paper indicates that black people who did not re-offend were twice as likely to be classified as high risk compared to a white person as the risk score assessment in Fig. 4 indicates.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
What the sources say
- Not checked yetcites [2] Angwin, J., et al.: (2016) Machine bias: there’s software used across the country to predict future criminals. And it’s biased against blacks. ProPublica. https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing (2016). Accessed 28 Mar 2021
hedgedcorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
None of the sources cited for this claim have been checked yet.
In the paper
“The algorithm applied the patterns in selecting individuals from the last 10 years, and it simply replicated that.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
What the sources say
- Not checked yetcites [33] Dastin, J.: Amazon scraps secret AI recruiting tool that showed bias against women. Reuters. https://www.reuters.com/article/us-amazon-com-jobs-automation-insight/amazon-scraps-secret-ai-recruiting-tool-that-showed-bias-against-women-idUSKCN1MK08G (2018). Accessed 24 Apr 2021 source
- Not checked yetcites [91] Vincent, J.: Amazon reportedly scraps internal AI recruiting tool that was biased against women. The Verge. https://www.theverge.com/2018/10/10/17958784/ai-recruiting-tool-bias-amazon-report (2018). Accessed 28 Mar 2021
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Defendants waiting for a trial with a high-risk score are more likely to be imprisoned while waiting for trial than those with low risks, so the consequences of a wrong assessment can be dire.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
hedgedcorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“Their business model operates on the basis of more prisoners, more profit [24].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
What the sources say
- Not checked yetcites [24] Burkhardt, B.C.: Who is in private prisons? Demographic profiles of prisoners and workers in American private prisons. Int. J. Law Crime Just. 51, 24–33 (2017) source
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“It predicts how likely the defendant is to re-offend by perpetrating a violent crime, taking into account up to 137 variables [61] such as gender and age and criminal history, with a specific weight given to each.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
What the sources say
- Not checked yetcites [61] Northpointe Inc.: Measurement & treatment implications of COMPAS core scales. Technical report, Northpointe Inc. https://www.michigan.gov/documents/corrections/Timothy Brenne Ph.D. Meaning and treatment implications of COMPA core scales 297495 7.pdf. Accessed 2 Feb 2020 (2009)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Following those patterns, Amazon’s system learns how to reinforce those normalised discriminatory outcomes.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“Finally, this model does not fully explain how it calculated those risk scores, so explicability seems non-existent [2].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
What the sources say
- Not checked yetcites [2] Angwin, J., et al.: (2016) Machine bias: there’s software used across the country to predict future criminals. And it’s biased against blacks. ProPublica. https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing (2016). Accessed 28 Mar 2021
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Minorities, especially from the Black community, seem to be the victims of racial injustice.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
hedgedcorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“However, a study of defendants in Broward County, Florida, showed that black individuals are much more likely to be classified as high risk [2].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
What the sources say
- Not checked yetcites [2] Angwin, J., et al.: (2016) Machine bias: there’s software used across the country to predict future criminals. And it’s biased against blacks. ProPublica. https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing (2016). Accessed 28 Mar 2021
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The private sector has an incentive to encourage incarcerating as many people from lower class backgrounds with restricted access to lawyers who are less likely to legally challenge unfair treatment.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“Correctional Offender Management Profiling for Alternative Sanction, known as COMPAS, is a predictive ML model designed to provide US courts with defendants recidivism risks scores that oscillate between 0 and 10.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
What the sources say
- Not checked yetcites [61] Northpointe Inc.: Measurement & treatment implications of COMPAS core scales. Technical report, Northpointe Inc. https://www.michigan.gov/documents/corrections/Timothy Brenne Ph.D. Meaning and treatment implications of COMPA core scales 297495 7.pdf. Accessed 2 Feb 2020 (2009)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
None of the sources cited for this claim have been checked yet.
In the paper
“One strategy was very straightforward by penalising the word woman. Any CV that contained this word or any others denoting the female gender of the applicant, like attending a women’s college, for example, downgraded the score, according to people familiar with the project [33, 91].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
What the sources say
- Not checked yetcites [33] Dastin, J.: Amazon scraps secret AI recruiting tool that showed bias against women. Reuters. https://www.reuters.com/article/us-amazon-com-jobs-automation-insight/amazon-scraps-secret-ai-recruiting-tool-that-showed-bias-against-women-idUSKCN1MK08G (2018). Accessed 24 Apr 2021 source
- Not checked yetcites [91] Vincent, J.: Amazon reportedly scraps internal AI recruiting tool that was biased against women. The Verge. https://www.theverge.com/2018/10/10/17958784/ai-recruiting-tool-bias-amazon-report (2018). Accessed 28 Mar 2021
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“For example, [86], provide the FairTest, a tool designed to test the ML model, assisting developers in checking data-driven applications to detect unfairness in outcomes.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > Blind testing
What the sources say
- Not checked yetcites [86] Tramer, F., Atlidakis, V., Geambasu, R., Hsu, D., Hubaux, J.P., Humbert, M., Juels, A., Lin, H.: FairTest: discovering unwarranted associations in data-driven applications. In: 2017 IEEE European Symposium on Security and Privacy (EuroS&P), pp. 401–416. IEEE (2017) source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In those cases, the protected groups’ data are unreliable, leading to an unreliable outcome.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > A more effective way to apply the protected groups’ concept
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Unfortunately, these safety measures do not seem to exist in the AI industry.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: unclear — It is unclear why the paper says it holds.
In the paper
“The participants may be withholding or providing false information to avoid retribution.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > A more effective way to apply the protected groups’ concept
What the sources say
- Not checked yetcites [94] Wang, S., Guo, W., Narasimhan, H., Cotter, A., Gupta, M., Jordan, M.I.: Robust optimization for fairness with noisy protected groups. arXiv:2002.09343 (2020)
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Moreover, there was a hope that ML models would remove those prejudices as machines executed the process, but the two case studies demonstrate that the opposite is true.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > A more effective way to apply the protected groups’ concept
boosteddirectn = 2- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“The benefit of AI is that most trials can be conducted in a virtual setting protecting the population from harm [53].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > Boxing methods
What the sources say
- Not checked yetcites [53] McDuff, D., Cheng, R., Kapoor, A.: Identifying bias in AI using simulation. arXiv:1810.00471 (2018)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
None of the sources cited for this claim have been checked yet.
In the paper
“There is a tendency in a male-dominated industry for some members of the staff to be apprehensive of a more gender-balanced working group [5, 75].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
What the sources say
- Not checked yetcites [5] Bagilhole, B.: Being different is a very difficult row to hoe: survival strategies of women academics. In: Davies, S., Lubelska, C., Quinn, J. (eds.) Changing the Subject, pp. 15–28. Taylor & Francis, London (2017) source
- Not checked yetcites [75] Sax, L.J., Lehman, K.J., Jacobs, J.A., Kanny, M.A., Lim, G., Monje-Paulson, L., Zimmerman, H.B.: Anatomy of an enduring gender gap: the evolution of women’s participation in computer science. J. Higher Educ. 88(2), 258–293 (2017) source
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Finally, there is no explicability of why two candidates with the same characteristics, except gender, are not invited to the interview.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: unclear — It is unclear why the paper says it holds.
In the paper
“Pharmaceutical companies test their products on different groups based on gender, ethnicity and age, amongst other factors, to unearth unexpected side effects that might not affect other groups of the population [77].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > Blind testing
What the sources say
- Not checked yetcites [77] Sedgwick, P.: Phases of clinical trials. BMJ 343, d6068 (2011) source
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
None of the sources cited for this claim have been checked yet.
In the paper
“Every time a new medicine is launched into the market, an independent body like the US FDA or the UK HMRA makes sure it is safe, or as safe as possible, and follows a clear set of guidelines [39, 43, 74, 77].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > An independent regulatory body as transnational as possible
What the sources say
- Not checked yetcites [39] Fleming, J.G.: Drug injury compensation plans. Am. J. Comp. Law. 1, 297–323 (1982) source
- Not checked yetcites [43] Hauben, M., Bate, A.: Decision support methods for the detection of adverse events in post-marketing data. Drug Discov. Today 14(7–8), 343–357 (2009) source
- Not checked yetcites [74] Santoro, M.A., Gorrie, T.M.: Ethics and the Pharmaceutical Industry. Cambridge University Press, Cambridge (2005) source
- Not checked yetcites [77] Sedgwick, P.: Phases of clinical trials. BMJ 343, d6068 (2011) source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“An error in an ML model can easily affect several countries in the case of Amazon or Facebook.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > An independent regulatory body as transnational as possible
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“For example, Google images were tagging images of black people with offensive, racist and incorrect remarks, on some occasions as gorillas [23].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > Blind testing
What the sources say
- Not checked yetcites [23] Buolamwini, J., Gebru, T.: Gender shades: intersectional accuracy disparities in commercial gender classification. In: Conference on fairness, accountability and transparency, pp. 77–91. PMLR (2018)
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“In a healthcare system that invests less money in black patients than white patients, the ML model feeding on that data can conclude that black patients are healthier than equally sick white patients and reduce the budget accordingly [62].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > Blind testing
What the sources say
- Not checked yetcites [62] Obermeyer, Z., Powers, B., Vogeli, C., Mullainathan, S.: Dissecting racial bias in an algorithm used to manage the health of populations. Science 366(6464), 447–453 (2019) source
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“For example, when Tramer et al. [86] tested FairTest to Healthcare Prediction, a winning approach from the Heritage Health Competition, they found out that there were errors when profiling some members of the elderly population, especially with some pre-existing health conditions [86, p. 4].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > Blind testing
What the sources say
- Not checked yetcites [86, p. 4]
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
None of the sources cited for this claim have been checked yet.
In the paper
“A bias-free AI is not achievable without an institution with enough power to guarantee compliance with those guidelines [26, 37, 70].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > Blind testing
What the sources say
- Not checked yetcites [26] Campolo, A., Sanfilippo, M., Whittaker, M., Crawford, K.: AI now 2017 report. https://assets.ctfassets.net/8wprhhvnpfc0/1A9c3ZTCZa2KEYM64Wsc2a/8636557c5fb14f2b74b2be64c3ce0c78/_AI_Now_Institute_2017_Report_.pdf (2017). Accessed 7 May 2021
- Not checked yetcites [37] Erdélyi, O. J., Goldsmith, J.: Regulating artificial intelligence proposal for a global solution. Preprint at arXiv:2005.11072 (2020)
- Not checked yetcites [70] Reisman, D., Schultz, J., Crawford, K., Whittaker, M.: Algorithmic Impact Assessments: A Practical Framework for Public Agency Accountability, pp. 1–22. AI Now Institute (2018)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
None of the sources cited for this claim have been checked yet.
In the paper
“It tends to lead to nothing, as Hagendorff [42, p. 108] asserts: “In their AI Now 2017 Report, Kate Crawford and her team state that ethics and forms of soft governance “face real challenges” [26, p. 5].”
- Also stated in: § An independent regulatory body as transnational as possible · plain
“This is mainly due to the fact that ethics has no enforcement mechanisms reaching beyond a voluntary and non-binding cooperation between ethicists and individuals working in research and industry”.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > An independent regulatory body as transnational as possible
What the sources say
- Not checked yetcites [42] Hagendorff, T.: The ethics of AI ethics: an evaluation of guidelines. Mind. Mach. 30(1), 99–120 (2020) source
- Not checked yetcites [26] Campolo, A., Sanfilippo, M., Whittaker, M., Crawford, K.: AI now 2017 report. https://assets.ctfassets.net/8wprhhvnpfc0/1A9c3ZTCZa2KEYM64Wsc2a/8636557c5fb14f2b74b2be64c3ce0c78/_AI_Now_Institute_2017_Report_.pdf (2017). Accessed 7 May 2021
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § An independent regulatory body as transnational as possible · plain
In the paper
“The UA framework is the primary tool to discover and analyse bias associated with the data used to train the algorithm.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > Blind testing
What the sources say
- Not checked yetcites [86] Tramer, F., Atlidakis, V., Geambasu, R., Hsu, D., Hubaux, J.P., Humbert, M., Juels, A., Lin, H.: FairTest: discovering unwarranted associations in data-driven applications. In: 2017 IEEE European Symposium on Security and Privacy (EuroS&P), pp. 401–416. IEEE (2017) source
boosteddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“As there is no satisfactory explanatory factor, then it is an example of an unwarranted association.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > Blind testing
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“Many laws and regulations have accelerated the fight for corruption, fiscal stability, accountability, Human Rights and fairness in members countries less willing to do so [47].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > An independent regulatory body as transnational as possible
What the sources say
- Not checked yetcites [47] Hoofnagle, C.J., van der Sloot, B., Borgesius, F.Z.: The European Union general data protection regulation: what it is and what it means. Inf. Commun. Technol. Law 28(1), 65–98 (2019) source
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Probably not a substantial claim. 56
- It may state two separate claims as one. 54
In the paper
“There is always a trade-off between accuracy, fairness and privacy that needs to be taken into account as Corbett-Davies et al. [31] examined in their study.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data
What the sources say
- Not checked yetcites [31] Corbett-Davies, S., Pierson, E., Feller, A., Goel, S., Huq, A.: Algorithmic decision making and the cost of fairness. In: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 797–806 (2017) source
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The machines incorporate those prejudices, becoming a proxy of humans and delegating responsibility.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“Humans can be biased as it is the way society is constructed and maintained by a minority elite at the top of the hierarchy.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data
What the sources say
- Not checked yetcites [17] Bonilla-Silva, E.: White Supremacy and Racism in the Post-Civil Rights Era. Lynne Rienner Publishers, Boulder (2001) source
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it.
In the paper
“Machine-learning models can have the capacity to evolve, develop and adapt their production in accordance with training information streams [3].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Definition of artificial intelligence, machine learning, algorithms and AI bias
What the sources say
- Not checked yetcites [3] Alpaydin, E. (2020). Introduction to machine learning. MIT press.
plainconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“While there are many reasons for incomplete or biased data, two are particularly relevant: historical human biases and incomplete or unrepresentative data [51].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data
What the sources say
- Not checked yetcites [51] Lee, N.T., Resnick, P., Barton, G.: Algorithmic bias detection and mitigation: best practices and policies to reduce consumer harms. Brookings Institute, Washington, DC (2019)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The bias impact assessment can be very helpful in clearly identifying the main stakeholders, their interests and their position of power when blocking or allowing necessary changes and the short- and long-term impacts.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > What a bias impact assessment is and how to develop one
boosteddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
None of the sources cited for this claim have been checked yet.
In the paper
“On the other hand, it is the approach by data scientists and programmers that characterise AI biases as bugs implying that it is just a technical issue like security that needs to be fixed [Tramer et al. 2016, 53, 54].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Introduction
What the sources say
- Not checked yetcites Tramer et al. 2016
- Not checked yetcites [53] McDuff, D., Cheng, R., Kapoor, A.: Identifying bias in AI using simulation. arXiv:1810.00471 (2018)
- Not checked yetcites [54] Mehrabi, N., Morstatter, F., Saxena, N., Lerman, K., Galstyan, A.: A survey on bias and fairness in machine learning. ACM Comput. Surv. 54(6), 1–35 (2021) source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
In the paper
“Of course, the first challenge in asking an algorithm to be fair or private is agreeing on what those words should mean in the first place—and not in the way a lawyer or philosopher might describe them, but in so precise a manner that they can be “explained” to a machine”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Introduction
What the sources say
- Not checked yetcites [49, p. 21]
plainconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
In the paper
“A set of prejudicial examples from which the model learns or in the case of under-represented groups which receives an incorrect or unfair valuation.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data
What the sources say
- Not checked yetcites [84] Thelisson, E., Padh, K., Celis, L.E.: Regulatory mechanisms and algorithms towards trust in AI/ML. In: Proceedings of the IJCAI 2017 Workshop on Explainable Artificial Intelligence (XAI), Melbourne, Australia (2017)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“For example, a lack geographical diversity in datasets like ImageNet (a large visual database designed for use in visual object recognition software research such as facial recognition) is an example for this type of bias [81]. This demonstrates a better representation of the pale skin population in the Western countries.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Definition of artificial intelligence, machine learning, algorithms and AI bias
What the sources say
- Not checked yetcites [81] Suresh, H., Guttag, J.V.: A framework for understanding unintended consequences of machine learning. arXiv:1901.10002 (2019)
boostedcorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“In addition, they have adopted an independent regulatory body (US FDA, UK MHRA or EU MDA)—a necessity that keeps coming up in many AI Ethics discussions [36].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Introduction
What the sources say
- Not checked yetcites [36] Erdélyi, O.J., Goldsmith, J.: Regulating Artificial Intelligence: Proposal for a Global Solution. In: Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society (2018) source
boosted- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Some data from specific groups’ databases can be either non-existent or simply incorrect, as was initially the case with female car drivers who were a minority.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“When the first safety belts and airbags for cars were designed, they suited tall males (the average engineer). Any other humans with other physical characteristics, especially shorter stature, were not considered, ending in a higher fatality rate in a car crash [18].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data
What the sources say
- Not checked yetcites [18] Bose, D., Segui-Gomez, S.C.D.M., Crandall, J.R.: Vulnerability of female drivers involved in motor vehicle crashes: an analysis of US population at risk. Am. J. Public Health 101(12), 2368–2373 (2011) source
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“Societies, as described by Bonilla-Silva [17], are structured by an elite at the high end of the hierarchy, controlling power and the top stages in the decision-making processes (e.g. judges, senior civil servants and politicians), whose biases have the ability to be adopted as a standard across society which then lead to historical human biases.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data
What the sources say
- Not checked yetcites [17] Bonilla-Silva, E.: White Supremacy and Racism in the Post-Civil Rights Era. Lynne Rienner Publishers, Boulder (2001) source
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
None of the sources cited for this claim have been checked yet.
In the paper
“Prisons and police enforcement seems to be a tool to perpetuate classism and maintain a rigid social structure, of which racism is a by-product [17, 22].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
What the sources say
- Not checked yetcites [17] Bonilla-Silva, E.: White Supremacy and Racism in the Post-Civil Rights Era. Lynne Rienner Publishers, Boulder (2001) source
- Not checked yetcites [22] Brewer, R.M., Heitzeg, N.A.: The racialization of crime and punishment: criminal justice, color-blind racism, and the political economy of the prison industrial complex. Am. Behav. Sci. 51(5), 625–644 (2008) source
hedgeddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“However, according to the statistics, it is evident that it is a socio-economic issue in which race plays a big part.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
boostedcorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“Black adults make up 33% of US prison population while just making 12% of the US adult population [102].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
What the sources say
- Not checked yetcites [102] Zeng, Z.: Jail Inmates in 2018, US Census Bureau, Bureau of Justice Statistics. https://bjs.ojp.gov/library/publications/jail-inmates-2018. Accessed 22 June 2021 (2020)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“Female programmers in IT constitute only 19.8% of the total workforce [79, p. 5]”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
What the sources say
- Not checked yetcites [79, p. 5]
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
None of the sources cited for this claim have been checked yet.
In the paper
“Although it was removed as soon as the bias was detected, and the company states that it was never in use, it is a good illustration of gender discriminatory outcomes as the case study will demonstrate [33, 91].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
What the sources say
- Not checked yetcites [33] Dastin, J.: Amazon scraps secret AI recruiting tool that showed bias against women. Reuters. https://www.reuters.com/article/us-amazon-com-jobs-automation-insight/amazon-scraps-secret-ai-recruiting-tool-that-showed-bias-against-women-idUSKCN1MK08G (2018). Accessed 24 Apr 2021 source
- Not checked yetcites [91] Vincent, J.: Amazon reportedly scraps internal AI recruiting tool that was biased against women. The Verge. https://www.theverge.com/2018/10/10/17958784/ai-recruiting-tool-bias-amazon-report (2018). Accessed 28 Mar 2021
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“For example, an individual who scores 8 has twice the re-offending rate of those who have 4.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“It is a problematic finding because Amazon has a long history in the use of algorithmic decision-making, as users have long been recommended products based on previous searches and purchases [52].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
What the sources say
- Not checked yetcites [52] Linden, G., Smith, B., York, J.: Amazon.com recommendations: Item-to-item collaborative filtering. IEEE Internet Comput. 7(1), 76–80 (2003) source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
In the paper
“Prison institutions in the US—private prison facilities, including non-secure community corrections centres and home confinement, held 15% of the federal prison population on December 31, 2017 [21, p. 16].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
What the sources say
- Not checked yetcites [21, p. 16]
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“There is a tension between fairness, one of the key values, and the tendency to maintain the status quo, which might be based on generational held prejudices against other groups, according to Bell [8].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
What the sources say
- Not checked yetcites [8] Bell, D. Faces at the Bottom of the Well: The Permanence of Racism. Hachette, UK (2018)
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
None of the sources cited for this claim have been checked yet.
In the paper
“The model ended in penalising the word woman and favouring male applicants.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
What the sources say
- Not checked yetcites [33] Dastin, J.: Amazon scraps secret AI recruiting tool that showed bias against women. Reuters. https://www.reuters.com/article/us-amazon-com-jobs-automation-insight/amazon-scraps-secret-ai-recruiting-tool-that-showed-bias-against-women-idUSKCN1MK08G (2018). Accessed 24 Apr 2021 source
- Not checked yetcites [91] Vincent, J.: Amazon reportedly scraps internal AI recruiting tool that was biased against women. The Verge. https://www.theverge.com/2018/10/10/17958784/ai-recruiting-tool-bias-amazon-report (2018). Accessed 28 Mar 2021
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“It can then help with how to mitigate them in the assessment of automated decision AI systems and facilitates accountability.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > What a bias impact assessment is and how to develop one
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The bias impact assessment does that and hence its relevance.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > What a bias impact assessment is and how to develop one
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The ML model COMPAS collects historical data from previous discriminatory court sentences and enhances those prejudices, with the added characteristic of being a proxy of a human and delegating moral responsibility.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: unclear — It is unclear why the paper says it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The assessment can also provide a better understanding of complex social and historical contexts as well as supporting a practical and ethical framework for spotting possible bias.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > What a bias impact assessment is and how to develop one
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it. It may state two separate claims as one.
In the paper
“There seems to be an imbalance of power between a privileged white population that holds a majority of high-ranking positions in the Justice and Police departments, which could favour institutionalised racism over the black population,8 as figures seem to demonstrate in the case of the black population being over-represented in the prison population [22].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
What the sources say
- Not checked yetcites [22] Brewer, R.M., Heitzeg, N.A.: The racialization of crime and punishment: criminal justice, color-blind racism, and the political economy of the prison industrial complex. Am. Behav. Sci. 51(5), 625–644 (2008) source
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The public sector, operating state prisons, seems to be willing to maintain the status quo by the figures provided in point 5.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
In the paper
“COMPAS is a risk-assessment tool that aids the operations of criminal justice organisations and is an extension of other judicial information systems [2].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
What the sources say
- Not checked yetcites [2] Angwin, J., et al.: (2016) Machine bias: there’s software used across the country to predict future criminals. And it’s biased against blacks. ProPublica. https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing (2016). Accessed 28 Mar 2021
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“The main reason is that Spiekermann’s model follows clear steps in identifying key elements such as stakeholders, benefits and harms while being complemented by the four value approach from the High-Level Expert EU paper.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > What a bias impact assessment is and how to develop one
boosteddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may state two separate claims as one.
None of the sources cited for this claim have been checked yet.
In the paper
“Reisman et al. [70, pp. 5–6] demonstrate the necessity of algorithmic impact assessment in AI as standard practice”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > What a bias impact assessment is and how to develop one
What the sources say
- Not checked yetcites [70, pp. 5–6]
- Not checked yetcites [25] Calvo, R.A., Peters, D., Cave, S.: Advancing impact assessment for intelligent systems. Nat. Mach. Intell. 2(2), 89–91 (2020) source
boosted- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“According to the World Prison Brief [100], the US has one of the highest incarceration rates in the world. In 2018, the figure was 23.4% of the total population.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
What the sources say
- Not checked yetcites [100] World Prison Brief.: https://prisonstudies.org/country/united-states-america (2018). Accessed 22 Apr 2021
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
In the paper
“and make up only a quarter of employees in the technology industry [82], p. 1]”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
What the sources say
- Not checked yetcites [82], p. 1]
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“The elite have the benefit of reinforcing privileges, and the rest of the population have limited access to progress to well-paid jobs and colleges [8].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
What the sources say
- Not checked yetcites [8] Bell, D. Faces at the Bottom of the Well: The Permanence of Racism. Hachette, UK (2018)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Someone can be wrongly imprisoned while awaiting trial who would not re-offend while a more dangerous individual more likely to offend would be let free (Fig. 3).”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“These charts show that scores for white defendants tended toward lower-risk categories. Scores for black defendants did not.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on race
What the sources say
- Not checked yetcites [2] Angwin, J., et al.: (2016) Machine bias: there’s software used across the country to predict future criminals. And it’s biased against blacks. ProPublica. https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing (2016). Accessed 28 Mar 2021
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“The concept fairness through awareness was introduced by Dwork et al. [35], and it states that if we wish to mitigate or remove bias in an algorithmic model, the first step is to be aware of the biases and why they occur.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > What a bias impact assessment is and how to develop one
What the sources say
- Not checked yetcites [35] Dwork, C., Hardt, M., Pitassi, T., Reingold, O., Zemel, R.: Fairness through awareness. In Proceedings of the 3rd Innovations in Theoretical Computer Science Conference, pp. 214–226 (2012) source
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“It proved challenging to remove bias and guarantee equal opportunities, so Amazon decided to scrap it altogether [33].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
What the sources say
- Not checked yetcites [33] Dastin, J.: Amazon scraps secret AI recruiting tool that showed bias against women. Reuters. https://www.reuters.com/article/us-amazon-com-jobs-automation-insight/amazon-scraps-secret-ai-recruiting-tool-that-showed-bias-against-women-idUSKCN1MK08G (2018). Accessed 24 Apr 2021 source
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“Tramer et al. [86, pp. 1 and 2] describe their Test themselves and its use as: “We report on the use of FairTest to investigate and in some cases address disparate impact, offensive labelling, and uneven rates of algorithmic error in four data-driven applications”.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > Blind testing
What the sources say
- Not checked yetcites [86, pp. 1 and 2]
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“For example, DuPont, a highly respected company that produced the popular material Teflon used in cooking tools, caused environmental damage that ended up costing the company around a billion dollars [78, p. 1].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry
What the sources say
- Not checked yetcites [78, p. 1]
hedgeddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In this specific example, the disadvantaged group are the women,9 as they are blocked or impeded from accessing those jobs and limiting respect for human autonomy (financial independence).”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The data used in training is intrinsically biased, as society is, and it tends to simply replicate human behaviour—one of the core aims of Artificial Intelligence.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > A more effective way to apply the protected groups’ concept
plaindirectn = 2- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Not doing the right thing can end up harming a business.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry
plainconditional- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it.
In the paper
“The AI industry is similar to the pharmaceutical industry in its multidisciplinary environment, its need for diverse voices and expertise, and the pivotal role that applied Ethics has played in its development.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry
What the sources say
- Not checked yetcites [74] Santoro, M.A., Gorrie, T.M.: Ethics and the Pharmaceutical Industry. Cambridge University Press, Cambridge (2005) source
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Thus consciously, or unconsciously, affecting the selection process.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“Although Amazon said it was never implemented, it did not confirm that recruiters had no access to the machine’s recommendations.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
What the sources say
- Not checked yetcites [33] Dastin, J.: Amazon scraps secret AI recruiting tool that showed bias against women. Reuters. https://www.reuters.com/article/us-amazon-com-jobs-automation-insight/amazon-scraps-secret-ai-recruiting-tool-that-showed-bias-against-women-idUSKCN1MK08G (2018). Accessed 24 Apr 2021 source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“A diverse team benefits the company in the sense that all voices are represented and their needs tailored [59].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
What the sources say
- Not checked yetcites [59] Nielsen, M.W., Alegria, S., Börjeson, L., Etzkowitz, H., Falk-Krzesinski, H.J., Joshi, A., Leahey, E., Smith-Doerr, L., Woolley, A.W., Schiebinger, L.: Opinion: gender diversity leads to better science. Proc. Natl. Acad. Sci. 114(8), 1740–1742 (2017) source
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it. It may state two separate claims as one.
In the paper
“That does not mean that there are no public safety issues like the anti-inflammatory drug Vioxx produced by Merck, which was linked to heart attacks and strokes in long-term use and withdrawn from the market in 2004 [10, 74, p. 13].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry
What the sources say
- Not checked yetcites [10, 74, p. 13]
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
In the paper
“A more effective use of this concept is tested by Wang et al. [94] in their paper when introducing the idea of noisy protected groups. By “noisy protected groups”, the authors mean when data are corrupted, missing or unreliable due to social pressures.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > A more effective way to apply the protected groups’ concept
What the sources say
- Not checked yetcites [94] Wang, S., Guo, W., Narasimhan, H., Cotter, A., Gupta, M., Jordan, M.I.: Robust optimization for fairness with noisy protected groups. arXiv:2002.09343 (2020)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“There were some attempts to ameliorate the problem, but the issue of gender discrimination was deep-rooted.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
What the sources say
- Not checked yetcites [33] Dastin, J.: Amazon scraps secret AI recruiting tool that showed bias against women. Reuters. https://www.reuters.com/article/us-amazon-com-jobs-automation-insight/amazon-scraps-secret-ai-recruiting-tool-that-showed-bias-against-women-idUSKCN1MK08G (2018). Accessed 24 Apr 2021 source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“As Sedgwick [77, p. 1] assures: “Drugs under development that are found to be unsafe or ineffective will not progress through all four phases”.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > Boxing methods
What the sources say
- Not checked yetcites [77, p. 1]
boosteddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The AI industry seems to enjoy a similar path of unregulated growth followed, hopefully, by regulations, better awareness in the industry and by the mainstream market—although it is vying to achieve this in a shorter period.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it. It may state two separate claims as one.
None of the sources cited for this claim have been checked yet.
In the paper
“The tech industry is male dominated, comprising almost 75% of the workforce [79, 82].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
What the sources say
- Not checked yetcites [79] Shields, M.: Women's participation in Seattle's high-tech economy. https://smartech.gatech.edu/bitstream/handle/1853/53790/madelyn_shields_womens_participation_in_seattles_hightech_economy.pdf (2015). Accessed 15 Aug 2021
- Not checked yetcites [82] Swift, S.: Gender Disparities in the Tech Industry: The Effects of Gender and Stereotypicability on Perceived Environmental Fit. In: 2015 NCUR (2015)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Thirdly, a lack of fairness, if a candidate matching those requisites is not selected for a job interview because of their being the wrong gender.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“There does not appear to be any reasons why tech companies cannot conduct their businesses in a similar manner to mitigate bias.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > Boxing methods
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“It also has a variety of collecting and contrasting data strategies like randomised control trials, which can help improve bias impact assessments by unearthing unexpected outcomes.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry
plainconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
In the paper
“A less diverse source of ideas and backgrounds might result in poorer creativity and overall innovation in producing new products.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
What the sources say
- Not checked yetcites [59] Nielsen, M.W., Alegria, S., Börjeson, L., Etzkowitz, H., Falk-Krzesinski, H.J., Joshi, A., Leahey, E., Smith-Doerr, L., Woolley, A.W., Schiebinger, L.: Opinion: gender diversity leads to better science. Proc. Natl. Acad. Sci. 114(8), 1740–1742 (2017) source
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“The main intention, at this stage, is to detect different outcomes from different groups and determine whether there is a fair reasoning behind this or whether the outcome is discriminatory, as seen in the COMPAS case study.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > Blind testing
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may state two separate claims as one.
In the paper
“However, side effects are usually noticed and recorded with the assistance of surveillance and monitoring systems like Pharmacovigilance in phase IV (Hauben et al. 2009), and there is a regulatory procedure to act upon it.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry
What the sources say
- Not checked yetcites (Hauben et al. 2009)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“Pharmaceutical companies were not expected to conduct trials to demonstrate the safety and accuracy of their medical products until 1962. That year, the US Congress passed the Kefauver-Harris Amendments to the Food, Drug and Cosmetics act of 1938, and Europe followed suit soon afterwards [74, p. 12].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry
What the sources say
- Not checked yetcites [74, p. 12]
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In addition, there is no prevention of harm (self-esteem/self-value), if women apply for those positions and it stops their career development.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“In addition, they conduct their trials on different age, gender and other characteristics groups at different stages and compare results. This process helps to develop a standard methodology to maximise benefits and remove, or minimise, harm.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“For example, in 2015, it was found that Google image misidentified black persons as gorillas [41]”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > Boxing methods
What the sources say
- Not checked yetcites [41] Guynn, J.: Google photos labelled black people 'gorillas'. USA today. http://www.usatoday.com/story/tech/2015/07/01/google-apologizes-after-photos-identify-black-people-as-gorillas/29567465/ (2015). Accessed 15 Mar 2021
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“On the other hand, regulation does not have to affect innovation, for example, when pharmaceutical companies developed a COVID-19 vaccine utilising a cutting-edge technology, mRNA, in a record time (Kim 2020).”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry
What the sources say
- Not checked yetcites (Kim 2020)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“Some of the male workers would prefer business as usual.10 They might be prejudiced against women and prefer the perpetuation of those values, making some women feel unwelcome in tech companies.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
What the sources say
- Not checked yetcites [92] Wajcman, J.: Feminism Confronts Technology. Penn State Press, Pennsylvania (1991)
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
None of the sources cited for this claim have been checked yet.
In the paper
“Other male workers, and the rest of the female workers, would prefer a more gender-balanced company where everybody feels welcome and the management board reflects that gender diversity [59, 101].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
What the sources say
- Not checked yetcites [59] Nielsen, M.W., Alegria, S., Börjeson, L., Etzkowitz, H., Falk-Krzesinski, H.J., Joshi, A., Leahey, E., Smith-Doerr, L., Woolley, A.W., Schiebinger, L.: Opinion: gender diversity leads to better science. Proc. Natl. Acad. Sci. 114(8), 1740–1742 (2017) source
- Not checked yetcites [101] Yasser, Q.R., Al Mamun, A., Ahmed, I.: Corporate social responsibility and gender diversity: insights from Asia Pacific. Corp. Soc. Responsib. Environ. Manag. 24(3), 210–221 (2017) source
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
In the paper
“Both case studies conclude with the necessity of applying a bias impact assessment to raise awareness of any bias and its possible reasons before being implemented as demonstrated by Dwork et al. [35]—and its urgency.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Algorithmic decision-making that discriminates and the problem with data > Algorithmic decision-making that discriminates based on gender
What the sources say
- Not checked yetcites [35] Dwork, C., Hardt, M., Pitassi, T., Reingold, O., Zemel, R.: Fairness through awareness. In Proceedings of the 3rd Innovations in Theoretical Computer Science Conference, pp. 214–226 (2012) source
boosted- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may state two separate claims as one.
In the paper
“Today, clinical trials are the norm. The drugs are tested on humans only after they have undergone laboratory testing. This takes the form of a series of successive clinical trials known as phase I, phase II, phase III, and phase IV trials.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > Boxing methods
What the sources say
- Not checked yetcites [77] Sedgwick, P.: Phases of clinical trials. BMJ 343, d6068 (2011) source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
In the paper
“It is designed to investigate associations between application outcomes (such as prices or premiums) and sensitive user attributes (such as race or gender). Once detected, it provides debugging capabilities that help programmers solve the detected unfair effects.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > Blind testing
What the sources say
- Not checked yetcites [86] Tramer, F., Atlidakis, V., Geambasu, R., Hsu, D., Hubaux, J.P., Humbert, M., Juels, A., Lin, H.: FairTest: discovering unwarranted associations in data-driven applications. In: 2017 IEEE European Symposium on Security and Privacy (EuroS&P), pp. 401–416. IEEE (2017) source
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
In the paper
“Phase I trials are conducted to determine the safety of the product, phase II trials test whether the drug is effective, phase III trials will compare its effectiveness with the standard treatment available, and phase IV trials monitor any risks and benefits once the product is in the market.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > Boxing methods
What the sources say
- Not checked yetcites [77] Sedgwick, P.: Phases of clinical trials. BMJ 343, d6068 (2011) source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
In the paper
“AI can save lives, for example, in the case of cancer detection technologies in medical imaging [12].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > An independent regulatory body as transnational as possible
What the sources say
- Not checked yetcites [12] Bi, W.L., Hosny, A., Schabath, M.B., Giger, M.L., Birkbak, N.J., Mehrtash, A., Allison, T., Arnaout, O., Abbosh, C., Dunn, I.F., Mak, R.H.: Artificial intelligence in cancer imaging: clinical challenges and applications. CA Cancer J Clin 69(2), 127–157 (2019) source
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“As Crawford [32, pp. 117–118] warns us: “The reproduction of harmful ideas is particularly dangerous now that AI has moved from being an experimental discipline used only in laboratories to being tested at scale on millions of people”.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Conclusion
What the sources say
- Not checked yetcites [32] Crawford, K.: The Atlas of AI. Yale University Press (2021) source
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Finally, good regulation mitigates harm, and it is more cost-effective than trying to cover the costs of damages and fines, as the previous example of DuPont shows.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > An independent regulatory body as transnational as possible
plaindirectSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
None of the sources cited for this claim have been checked yet.
In the paper
“Studies by [8, 55, 67], are potent illustrators of embedded bias in society and the difficulty of removing them which is reflected in the AI industry.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Conclusion
What the sources say
- Not checked yetcites [8] Bell, D. Faces at the Bottom of the Well: The Permanence of Racism. Hachette, UK (2018)
- Not checked yetcites [55] Mills, C.W.: The Racial Contract. Cornell University Press, Ithaca (2014)
- Not checked yetcites [67] Pateman, C.: The Sexual Contract. Wiley, Weinheim (2018)
boostedcorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Third world countries poor in resources can rely on the seal of approval from rich countries to allow a medicine to reach their market.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > An independent regulatory body as transnational as possible
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
- Rests on an assumption: Regulatory approval from the FDA makes approval by other national agencies highly probable.
In the paper
“Elderly people with some pre-existing health conditions were discriminated against incorrectly, predicting more visits to the hospital, in some cases, and being charged a higher premium.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > Blind testing
What the sources say
- Not checked yetcites [86, p. 4]
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“An example can be when Gmail introduced an algorithm to remove spam emails [11]—the same principle could be used to reduce fake news.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > An independent regulatory body as transnational as possible
What the sources say
- Not checked yetcites [11] Bhuiyan, H., Ashiquzzaman, A., Juthi, T.I., Biswas, S., Ara, J.: A survey of existing e-mail spam filtering methods considering machine learning techniques. Glob. J. Comput. Sci. Technol. 18(2-c)(2018)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“as international companies devise complex accounting mechanisms to avoid paying much tax (as becoming more common in the EU and US, Bageri 2013).”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > An independent regulatory body as transnational as possible
What the sources say
- Not checked yetcites Bageri 2013 source
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“However, by its device-based nature, AI is a transnational technology, and it needs solutions applicable beyond borders.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > An independent regulatory body as transnational as possible
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
In the paper
“The association-guided tree construction algorithm further investigates the findings in the first step. It introduces a visual presentation to facilitate the interpretation and identify which subsets are affected by algorithmic bias [86].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > Blind testing
What the sources say
- Not checked yetcites [86] Tramer, F., Atlidakis, V., Geambasu, R., Hsu, D., Hubaux, J.P., Humbert, M., Juels, A., Lin, H.: FairTest: discovering unwarranted associations in data-driven applications. In: 2017 IEEE European Symposium on Security and Privacy (EuroS&P), pp. 401–416. IEEE (2017) source
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
In the paper
“As they detailed in the paper: the initial AIF360 Python package implements inputs from 8 published papers, over 71 bias detection metrics, and 9 bias mitigation algorithms.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > Blind testing
What the sources say
- Not checked yetcites [9] Bellamy, R.K., Dey, K., Hind, M., Hoffman, S.C., Houde, S., Kannan, K., Lohia, P., Martino, J., Mehta, S., Mojsilovic, A., Nagar, S. AI fairness 360: an extensible toolkit for detecting, understanding, and mitigating unwanted algorithmic bias. arXiv:1810.01943 (2018)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“Woolley et al. [99] provides case studies as examples such as: political bot intervention during pivotal events in Brazil or the origins of digital misinformation in Russia in which manipulation took place.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > An independent regulatory body as transnational as possible
What the sources say
- Not checked yetcites [99] Woolley, S.C., Howard, P.N. (eds.) Computational Propaganda: Political Parties, Politicians, and Political Manipulation on Social Media. Oxford University Press, Oxford (2018)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
In the paper
“In addition, tech companies like Google or Facebook operate beyond borders and its implications when things go wrong are transnational [99].”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Possible machine-centric solutions adapted from or inspired by the pharmaceutical industry > An independent regulatory body as transnational as possible
What the sources say
- Not checked yetcites [99] Woolley, S.C., Howard, P.N. (eds.) Computational Propaganda: Political Parties, Politicians, and Political Manipulation on Social Media. Oxford University Press, Oxford (2018)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“When scholars look back at discovering previous threats to humankind like climate change due to man-made pollution, as early as the 80s [90], they realise that providing the evidence is insufficient to modifying behaviour by companies and governments and they surmise that more pro-active strategies are needed.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Conclusion
What the sources say
- Not checked yetcites [90] Verdin, J., Funk, C., Senay, G., Choularton, R.: Climate science and famine early warning. Philos. Trans. R. Soc. B Biol. Sci. 360(1463), 2155–2168 (2005) source
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Third, instances of unfairness in one group might not translate into another group, as it is in the previously explained Simpson’s Paradox. For example, long waiting lists for a cancer treatment is considered unfair in the general population, but does not affect those with a private health insurance.”
§ AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry > Conclusion
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds. It may state two separate claims as one.
The paper relies on this without support.
The pharmaceutical industry has a long history of developing stage-by-stage risk-assessment methodologies for known and unknown risks.
“Such industries have a long history of developing risk-assessment methodologies, on a stage-by-stage basis, facing the known and the unknown.”
The paper relies on this without support.
The pharmaceutical industry has a long history of Applied Ethics.
“The pharmaceutical industry also has a long history of Applied Ethics,2 which will be explored.”
The paper relies on this without support.
The pharmaceutical industry has adopted independent regulatory bodies such as the US FDA, UK MHRA, or EU MDA.
“In addition, they have adopted an independent regulatory body (US FDA, UK MHRA or EU MDA)—a necessity that keeps coming up in many AI Ethics discussions [36].”
The paper relies on this without support.
Amazon is at the forefront of algorithmic decision-making technologies.
“AI has been at the heart of their business for years and they are hence assumed to be at the forefront of such technologies.”
The paper relies on this without support.
The pharmaceutical industry's long history of applied ethics and risk-assessment methods is advantageous for finding ethical solutions.
“The pharmaceutical industry has a long history in applied Ethics and risk-assessment methodologies in a multidisciplinary field, which is advantageous in finding ethical solutions.11”
The paper relies on this without support.
Current business models do not address equal-society and protected-group concerns and cannot be legally enforced.
“All ML models affect those issues in one way or another, and the current business models do not address these concerns with no way to legally enforce them.”
The paper relies on this without support.
No medicine can reach the market without a risk and safety report, regulatory approval, and follow-up monitoring.
“No medicine would be allowed to reach the market without a risk and safety report and seal of approval from the regulatory body and monitoring systems for a follow-up.”
The paper relies on this without support.
Most people worldwide have smartphones and, to a lesser degree, computers.
“The majority of the global population has smartphones, and to a limited extent, computers, the damage could be extended to the whole planet.”
The paper relies on this without support.
Algorithmic recommendations influence an overworked official's decision despite the judge having final say.
“Although the judge has the final say, algorithmic recommendations affect an overworked official’s decision.”
The paper relies on this without support.
Most people are not expert in or aware of these kinds of algorithmic harms.
“The majority of the population does not have the expertise or awareness in relation to these types of harms.”
The paper relies on this without support.
Independent pharmaceutical regulatory bodies have developed similar guidelines.
“Those independent bodies have developed similar guidelines.”
The paper relies on this without support.
Regulatory approval from the FDA makes approval by other national agencies highly probable.
“For example, a medicine approved by the FDA is very likely to be approved by other national bodies.”
The paper relies on this without support.
Soft-law or advisory approaches have been tried for years without much progress.
“The argument for a more advisory role, or soft law, has been practised for years with hardly any progress.”
The paper relies on this without support.
Advisory bodies often phrase recommendations vaguely so as to satisfy every party.
“Recommendations by advisory bodies tend to resort to vague language to accommodate all parties’ interests.”
The paper relies on this without support.
Current technology and awareness allow us at least to reduce bias, with the goal of eliminating it.
“We have now the technologies and awareness to at least mitigate, aiming to remove, bias.”
The paper relies on this without support.
AI is now globally pervasive and powerful while remaining opaque.
“AI has become so ubiquitous globally and powerful in a non-transparent way that setting up an independent body is an urgent necessity.”
The paper relies on this without support.
Fairness and bias are understood differently by different people and lack a single definition, though some core values may be shared.
“First, the concept of fairness and bias can mean different things for different people, it lacks uniformity although some basic principles, or values as previously described, can be agreed.”
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Bias exists in PD-ai prediction model due to societal and representation barriers that arising from geographical diversity in data collection [15].”
From Artificial intelligence bias in medical system designs: a systematic review (2023) · cites this paper as [15]
Citations
Cites
Artificial intelligence: A modern approach
Nilsson, N. J. (1996) · Artificial Intelligence · doi:10.1016/0004-3702(96)00007-0 · Cited by 10,730
(2012) · Cited by 9,062
Dissecting 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
A Survey on Bias and Fairness in Machine Learning
Mehrabi, N., Morstatter, F., Saxena, N., Lerman, K., & Galstyan, A. (2022) · ACM Computing Surveys · doi:10.1145/3457607 · Cited by 5,812 · cited 2×
“This a non-comprehensive list that shows some of the most common type of bias that needs to be dealt with [54] and Suresh et al.”
BackgroundAmazon.com recommendations: item-to-item collaborative filtering
Linden, G., Smith, B., & York, J. (2003) · IEEE Internet Computing · doi:10.1109/mic.2003.1167344 · Cited by 5,403
Dwivedi, Y. K., Hughes, L., Ismagilova, E., Aarts, G., Coombs, C., Crick, T., Duan, Y., Dwivedi, R., Edwards, J., Eirug, A., et al. (2021) · International Journal of Information Management · doi:10.1016/j.ijinfomgt.2019.08.002 · Cited by 4,611
Dwork, C., Hardt, M., Pitassi, T., Reingold, O., & Zemel, R. (2012) · Proceedings of the 3rd Innovations in Theoretical Computer Science Conference · doi:10.1145/2090236.2090255 · Cited by 3,501 · cited 2×
“[35], and it states that if we wish to mitigate or remove bias in an algorithmic model, the first step is to be aware of the biases and why they occur.”
BackgroundWeapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy
Doyle, T. (2017) · The Information Society · doi:10.1080/01972243.2017.1354593 · Cited by 3,111 · cited 1×
“Whilst useful in creating debate, these tend to present either no solutions at all or overly simplified single solutions [13, 15, 19, 29, 32, 35, 56, 60, 64, 89, 96, 104].”
BackgroundMachine learning and deep learning
Janiesch, C., Zschech, P., & Heinrich, K. (2021) · Electronic Markets · doi:10.1007/s12525-021-00475-2 · Cited by 2,959
Algorithms of Oppression: How Search Engines Reinforce Racism
Noble, S. U. (2020) · doi:10.2307/j.ctt1pwt9w5 · Cited by 2,709 · cited 4×
“These benchmarks can be used in the evaluation of facial recognition systems that were biased towards skin colour and gender [23, 60].”
Influential
Research works printed but not linked: 15
- Anderson, J., Rainie, L., Luchsinger, A.: Artificial intelligence and the future of humans. Pew Res. Center 10, 12 (2018)
- Barocas, S., Selbst, A.D.: Big data’s disparate impact. Calif. L. Rev. 104, 671 (2016)
- Binns, R.: Fairness in machine learning: lessons from political philosophy. In Conference on Fairness, Accountability and Transparency, pp. 149–159. PMLR (2018)
- Bronson, J., Carson, E.A.: Prisoners in 2017. Age 500, 400 (2019)
- Buolamwini, J., Gebru, T.: Gender shades: intersectional accuracy disparities in commercial gender classification. In: Conference on fairness, accountability and transparency, pp. 77–91. PMLR (2018)
- Goodfellow, I., Bengio, Y., Courville, A.: Deep learning. MIT Press, Cambridge (2016)
- Lee, N.T., Resnick, P., Barton, G.: Algorithmic bias detection and mitigation: best practices and policies to reduce consumer harms. Brookings Institute, Washington, DC (2019)
- Mills, C.W.: The Racial Contract. Cornell University Press, Ithaca (2014)
- Nabirahni, D.M., Evans, B.R., Persaud, A.: Al-Khwarizmi (algorithm) and the development of algebra. Math. Teach. Res. J. 11(1–2), 13–17 (2019)
- Pateman, C.: The Sexual Contract. Wiley, Weinheim (2018)
- Reisman, D., Schultz, J., Crawford, K., Whittaker, M.: Algorithmic Impact Assessments: A Practical Framework for Public Agency Accountability, pp. 1–22. AI Now Institute (2018)
- Thelisson, E., Padh, K., Celis, L.E.: Regulatory mechanisms and algorithms towards trust in AI/ML. In: Proceedings of the IJCAI 2017 Workshop on Explainable Artificial Intelligence (XAI), Melbourne, Australia (2017)
- Washington, A.L.: How to argue with an algorithm: lessons from the COMPAS-ProPublica debate. Colo. Tech. LJ 17, 131 (2018)
- Whittlestone, J., Nyrup, R., Alexandrova, A., Dihal, K., Cave, S.: Ethical and societal implications of algorithms, data, and artificial intelligence: a roadmap for research. Nuffield Foundation, London (2019)
- Woolley, S.C., Howard, P.N. (eds.) Computational Propaganda: Political Parties, Politicians, and Political Manipulation on Social Media. Oxford University Press, Oxford (2018)
Cited by
Sorted by citations
Chen, F., Wang, L., Hong, J., Jiang, J., & Zhou, L. (2024) · Journal of the American Medical Informatics Association · doi:10.1093/jamia/ocae060 · Cited by 178
Artificial intelligence in the field of pharmacy practice: A literature review
Chalasani, S. H., Syed, J. M., Ramesh, M., Patil, V., & Kumar, T. M. P. (2023) · Exploratory Research in Clinical and Social Pharmacy · doi:10.1016/j.rcsop.2023.100346 · Cited by 231
“98 Minorities, for example, may be under-represented in datasets due to racial biases in dataset creation, resulting in lower-than-expected prediction performance.”
BackgroundUnderstanding and Mitigating Bias in Imaging Artificial Intelligence
Tejani, A. S., Ng, Y. S., Xi, Y., & Rayan, J. C. (2024) · RadioGraphics · doi:10.1148/rg.230067 · Cited by 138
The future of pharmaceuticals: Artificial intelligence in drug discovery and development
Fu, C., & Chen, Q. (2025) · Journal of Pharmaceutical Analysis · doi:10.1016/j.jpha.2025.101248 · Cited by 151
“A predictive study in 2022 concluded that by heavily investing in AI, the pharmaceutical industry could see a return on investment increase of more than 45% [81].”
Ethical considerations on artificial intelligence in dentistry: A framework and checklist
Rokhshad, R., Ducret, M., Chaurasia, A., Karteva, T., Radenković, M., Roganović, J., Hamdan, M. H., Mohammad‐Rahimi, H., Krois, J., Lahoud, P., et al. (2023) · Journal of Dentistry · doi:10.1016/j.jdent.2023.104593 · Cited by 149
Metadata
- Authors
- Belenguer, L.
- Corresponding author
- Belenguer, Lorenzo
- ORCID
- Belenguer, Lorenzo
- Venue
- AI and Ethics
- Published
- 2022-02-10
- Volume
- 2
- Issue
- 4
- Pages
- 771-787
- ISSN
- 2730-5953 (ISSN-L)
- 2730-5961
- PMC
- PMC8830968
- OpenAlex
- W4211155626
- Semantic Scholar
- 2db0876940068279e829854a498337ddff796432
- Cited by
- 289
- References
- 70
- Citations by year
- 2026: 71 · 2025: 80 · 2024: 59 · 2023: 78 · 2022: 1
- FWCI
- 63.52
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Licence
- Locations
- AI and Ethics (publishedVersion)
- PubMed Central (publishedVersion)
- PubMed (publishedVersion)
- Funders
- No funding was received
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
- Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
- Social Sciences › Business, Management and Accounting › Organizational Behavior and Human Resource Management › AI and HR Technologies
- Keywords
- algorithmic bias, algorithmic decision-making, drug trials, social prejudice, independent oversight bodies, AI ethics framework
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.
A language model drew the claims from the paper on 2026-10-05, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-05 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
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
