The uselessness of AI ethics

Munn, L. The uselessness of AI ethics.

AI Ethics (2022).

https://doi.org/10.1007/s43681-022-00209-w

Abstract

As the awareness of AI’s power and danger has risen, the dominant response has been a turn to ethical principles. A flood of AI guidelines and codes of ethics have been released in both the public and private sector in the last several years. However, these are meaningless principles which are contested or incoherent, making them difficult to apply; they are isolated principles situated in an industry and education system which largely ignores ethics; and they are toothless principles which lack consequences and adhere to corporate agendas. For these reasons, I argue that AI ethical principles are useless, failing to mitigate the racial, social, and environmental damages of AI technologies in any meaningful sense. The result is a gap between high-minded principles and technological practice. Even when this gap is acknowledged and principles seek to be “operationalized,” the translation from complex social concepts to technical rulesets is non-trivial. In a zero-sum world, the dominant turn to AI principles is not just fruitless but a dangerous distraction, diverting immense financial and human resources away from potentially more effective activity. I conclude by highlighting alternative approaches to AI justice that go beyond ethical principles: thinking more broadly about systems of oppression and more narrowly about accuracy and auditing.

(cut)

Meaningless principles

The deluge of AI codes of ethics, frameworks, and guidelines in recent years has produced a corresponding raft of principles. Indeed, there are now regular meta-surveys which attempt to collate and summarize these principles. However, these principles are highly abstract and ambiguous, becoming incoherent. Mittelstadt suggests that work on AI ethics has largely produced “vague, high-level principles, and value statements which promise to be action-guiding, but in practice provide few specific recommendations and fail to address fundamental normative and political tensions embedded in key concepts.” The point here is not to debate the merits of any one value over another, but to highlight the fundamental lack of consensus around key terms. Commendable values like “fairness” and “privacy” break down when subjected to scrutiny, leading to disparate visions and deeply incompatible goals.

What are some common AI principles? Despite the mushrooming of ethical statements, Floridi and Cowls suggest many values recur frequently and can be condensed into five core principles: beneficence, non-maleficence, autonomy, justice, and explicability. These ideals sound wonderful. After all, who could be against beneficence? However, problems immediately arise when we start to define what beneficence means. In the Montreal principles for instance, “well-being” is the term used, suggesting that AI development should promote the “well-being of all sentient creatures.” While laudable, clearly there are tensions to consider here. We might think, for instance, of how information technologies support certain conceptions of human flourishing by enabling communication and business transactions—while simultaneously contributing to carbon emissions, environmental degradation, and the climate crisis. In other words, AI promotes the well-being of some creatures (humans) while actively undermining the well-being of others.

The same issue occurs with the Statement on Artificial Intelligence, Robotics, and Autonomous Systems. In this Statement, beneficence is gestured to through the concept of “sustainability,” asserting that AI must promote the basic preconditions for life on the planet. Few would argue directly against such a commendable aim. However, there are clearly wildly divergent views on how this goal should be achieved. Proponents of neoliberal interventions (free trade, globalization, deregulation) would argue that these interventions contribute to economic prosperity and in that sense sustain life on the planet. In fact, even the oil and gas industry champions the use of AI under the auspices of promoting sustainability. Sustainability, then, is a highly ambiguous or even intellectually empty term that is wrapped around disparate activities and ideologies. In a sense, sustainability can mean whatever you need it to mean. Indeed, even one of the members of the European group denounced the guidelines as “lukewarm” and “deliberately vague,” stating they “glossed over difficult problems” like explainability with rhetoric.

Moral dilemmas in (not) treating patients who feel they are a burden

Metselaar S, Widdershoven G.
[published online April 23, 2019]
Bioethics. 2019;33(4):431-438.

Abstract

Working as clinical ethicists in an academic hospital, we find that practitioners tend to take a principle‐based approach to moral dilemmas when it comes to (not) treating patients who feel like a burden, in which respect for autonomy tends to trump other principles. We argue that this approach insufficiently deals with the moral doubts of professionals with regard to feeling that you are a burden as a motive to decline or withdraw from treatment. Neither does it take into adequately account the specific needs of the patient that might underlie their feeling of being a burden to others. We propose a care ethics approach as an alternative. It focuses on being attentive and responsive to the caring needs of those involved in the care process—which can be much more specific than either receiving or withdrawing from treatment. This approach considers these needs in the context of the patient's identity, biography and relationships, and regards autonomy as relational rather than as individual. We illustrate the difference between these two approaches by means of the case of Mrs K. Furthermore, we show that a care ethics approach is in line with interventions that are found to alleviate feeling a burden and maintain that facilitating moral case deliberation among practitioners can supports them in taking a care ethics approach to moral dilemmas in (not) treating patients who feel like a burden.

The info is here.

Alleviating Burdensome Beliefs Through a Care Ethics Approach

Medical Bag
Originally posted May 29, 2019

Compared with a principles-based approach, taking a care ethics approach to patients who believe they are a burden may be more effective for addressing moral dilemmas related to treatment, according to research published in Bioethics.

Two clinical ethicists from the department of medical humanities at VU University Medical Center in Amsterdam, The Netherlands, shared the case of Mrs K, a 66-year-old patient with leukemia, and examined the ways in which physicians can approach treating a patient who feels like a burden.

Mrs K recently received a bone marrow transplant, but because of rejection symptoms, is now taking an antirejection treatment. Although a cure is possible, the treatment is both taxing and extensive and presents a host of physical and mental challenges. Although Mrs K had previously focused on survival, her mindset has shifted: She says that she is burdening her husband and feels that he deserves better. Mrs K feels that life is no longer worth living and has considered stopping her antirejection treatment, which will result in her death.

Noticing that Mrs K’s mood has been poor over a long period of time, the treating physician suggests antidepressant therapy; they believe that by treating the patient’s depression, the patient will be more optimistic about continuing the antirejection therapy. Mrs K’s husband — also a physician — strongly disagrees with this course of treatment. Mrs K’s care team contacts the clinical ethicist to address this moral dilemma.

The info is here.

Where did Nazi doctors learn their ethics? From a textbook

Michael Cook
BioEdge.org
Originally posted April 29, 2017

German medicine under Hitler resulted in so many horrors – eugenics, human experimentation, forced sterilization, involuntary euthanasia, mass murder – that there is a temptation to say that “Nazi doctors had no ethics”.

However, according to an article in the Annals of Internal Medicine by Florian Bruns and Tessa Chelouche (from Germany and Israel respectively), this was not the case at all. In fact, medical ethics was an important part of the medical curriculum between 1939 and 1945. Nazi officials established lectureships in every medical school in Germany for a subject called “Medical Law and Professional Studies” (MLPS).

There was no lack of ethics. It was just the wrong kind of ethics.

(cut)

It is important to realize that ethical reasoning can be corrupted and that teaching ethics is, in itself, no guarantee of the moral integrity of physicians.

The article is here.

Embedding Ethical Principles in Collective Decision Support Systems

Joshua Greene, Francesca Rossi, John Tasioulas, Kristen Brent Venable, & Brian Williams
Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence (AAAI-16)

Abstract

The future will see autonomous machines acting in the same environment as humans, in areas as diverse as driving, assistive technology, and health care. Think of self-driving cars, companion robots, and medical diagnosis support systems.  We also believe that humans and machines will often need to work together and agree on common decisions. Thus hybrid collective decision making systems will be in great need.  In this scenario, both machines and collective decision making systems should follow some form of moral values and ethical principles (appropriate to where they will act but always aligned to humans’), as well as safety constraints. In fact, humans would accept and trust more machines that behave as ethically as other humans in the same environment. Also, these principles would make it easier for machines to determine their actions and explain their behavior in terms understandable by humans. Moreover, often machines and humans will need to make decisions together, either through consensus or by reaching a compromise. This would be facilitated by shared moral values and ethical principles.

The article is here.

Ethical Principles and the Communication of Forensic Mental Health Assessments

Alfred Allan, Thomas Grisso
Ethics & Behavior 
Vol. 24, Iss. 6, 2014

Abstract
Our premise is that ethics is the essence of good forensic practice and that mental health professionals must adhere to the ethical principles, standards, and guidelines of their professional bodies when they communicate their findings and opinions. We demonstrate that adhering to ethical principles can improve the quality of forensic reports and communications. We demonstrate this by focusing on the most basic principles that underlie professional ethical standards and guidelines, namely, Fidelity and Responsibility, Integrity, Respecting Rights and Dignity of Persons, and Justice and Fairness. For each principle we offer a brief definition and explain its demands. Then we identify ways in which the principle can guide the organization, content, or style of forensic mental health report writing, offering illustrative examples that demonstrate or abuse the principle.

(cut)

Unless they are confronted with a specific ethical problem, many professionals consider the writing of reports an archetypical practical task and might not consider how it is related to the ethical principles of their profession. Yet ethics is the very essence of professional practice. As we seek to demonstrate, almost every facet of report writing is related to an ethical principle. We do not assert that one is practicing unethically if one makes the types of errors in report writing that we have described. Our primary purpose in this article is to demonstrate how forensic report writing can be improved by using professional ethics principles as a guide for report-writing practice.

The entire article is here, behind a paywall.

Are Human Rights Redundant in the Ethical Codes of Psychologists?

Alfred Allan
Ethics & Behavior
Volume 23, Issue 4, 2013
DOI:10.1080/10508422.2013.776480

The codes of ethics and conduct of a number of psychology bodies explicitly refer to human rights, and the American Psychological Association recently expanded the use of the construct when it amended standard 1.02 of the Ethical Principles of Psychologists and Code of Conduct. What is unclear is how these references to human rights should be interpreted. In this article I examine the historical development of human rights and associated constructs and the contemporary meaning of human rights. As human rights are generally associated with law, morality, or religion, I consider to which of forms of these references most likely refer. I conclude that these references in ethical codes are redundant and that it would be preferable not to refer to human rights in codes. Instead, the profession should acknowledge human rights as a separate and complimentary norm system that governs the behavior of psychologists and should ensure that they have adequate knowledge of human rights and encourage them to promote human rights.

The entire article is here.

The Limits of Moral Argument: Tamler Sommer presents at TEDx

Tamler Sommers is an Assistant Professor of Philosophy at the University of Houston and holds a joint appointment with the Honors College. His research and teaching are in the areas of ethics, political philosophy, and the philosophy of law, specializing in issues relating to free will, moral responsibility, punishment, and revenge.

Uploaded on December 17, 2011





The link to this video will be kept in the Audio and Video Resource page of this site.

Very Bad Wizards: Trolleys, Utilitarianism, and Psychopaths

Tamler Sommers

Very Bad Wizards Podcast

A philosopher and a psychologist ponder the nature of human morality

Published on October 20, 2012

Notes

Tamler contemplates ending it all because he can't get 'Call Me Maybe' out of his head, and Dave doesn't try to talk him out of it. This is followed by a discussion about drones, psychopaths, Canadians, Elle Fanning, horrible moral dilemmas, and the biggest rivalry in Ethics: utilitarians vs. Kantians.

Episode 6 page: Trolleys, Utilitarianism and Psychopaths

The podcast is here. 

Editorial notes: Very Bad Wizards is a series of podcasts that combine psychological and philosophical perspectives on a variety of topics.   In terms of informed consent, the language is rough and the humor is..........let's just say bawdy, crude and coarse.   Truly, the podcast is not for the faint of heart.

If you like this podcast, there are a variety of Very Bad Wizards podcasts.  While their brand of humor is part of the content, the episodes that I have found as potentially good teaching tools are found in the Audio and Video resource page of this site.

And, I have Tamler's picture on this page to help boost his self-esteem.  Apparently, he feels badly that his TEDx talk has less views that Dave Pizarro's TEDx talk.  Both of their TEDx  talks are also on the Audio and Video resource page.

Medical ethics language doesn’t stick with students

A study finds a gap between learning ethical terms and using them in a clinical setting, which can lead to a lack of shared understanding.

By MARCIA FRELLICK
amednews correspondent — Posted May 15, 2013

Do medical school students remember ethical principles when they start practicing medicine?

Because physicians need shared language and universal terms when they discuss ethical issues with each other and with patients, researchers at the University of Iowa Carver College of Medicine examined how well terms and concepts taught in school were recalled and incorporated in the clinical years. The study asked 109 third-year medical students at the University of Iowa to recall ethics terms learned in the first two years of school.

Results were mixed, according to the study posted online April 14 in AJOB Primary Research. The students were much more likely to name the four ethical principles (beneficence, nonmaleficence, respect for patient autonomy and justice) than the six sources of ethical value or categories for justifying an ethical decision (ethical principles, rights, consequences, comparable cases, professional guidelines and conscientious practice).

Overall, 59.6% of the students remembered all four principles, but the highest number of sources of ethical value recalled was four of the six. Only 10% of students could name three or four of the six sources.

The entire article is here.

Here is the original research abstract.

Background: 
Ethics education is an established part of the medical school curriculum and typically involves preclinical instruction that includes formal ethical terminology. However, it is not clear whether the language of ethics taught in preclinical settings is applied by students during the clinical years of training.

Methods: 
We used a survey and a content analysis of written reflections to determine whether third-year (clinical) medical students were able to recall and apply ethical principles and other sources of ethical value they were taught as second-year (preclinical) students.

Results: 
The majority of clinical students were able to recall the four ethical principles, appreciated the relevance of preclinical ethics education, and had positive self-assessments of their clinical-ethical reasoning abilities. However, they were less able to recall other (nonprinciple) sources of ethical value and infrequently used ethical terms spontaneously in written reflections about ethically or professionally challenging issues.

Conclusions: 
Ethics educators should consider the extent to which preclinical ethics education depends on a formal language of ethics and should develop ways to reinforce that language meaningfully through experience-based learning opportunities during the clinical years of training, with special emphasis on the way clear ethical reasoning and communication demonstrate respect for other persons.

Michael Shermer - The Moral Arc of Science

The Fiduciary Heart of Ethics

By Ed Zuckerman, PhD


We have all heard the word “fiduciary”bandied about during the ongoing mortgage and banking crisis and probably thought it meant something about money. Not true.

In your first day in professional school, when they talked about the helping relationship, you were probably told something like, always put your client’s needs ahead of your own. As soon as you nodded to that you became a fiduciary.

The fiduciary duty then is an obligation to act in the best interests of another party. In any relationship it is imposed upon the stronger (more knowledgeable, richer, smarter, whatever) as a duty to the weaker. Stealing candy from a baby is a breach of this duty and so is insider trading–using non-public information gained because of one’s special position to buy stock for a gain not available to the public. In one sense being a good parent is fulfilling a fiduciary obligation to a child - doing what is in the child’s best long-term interest instead of what is convenient for the parent. 

However, stockbrokers only have to offer what is suitable - likely to be beneficial -  to the client. They do not have to offer what might be in the clients best interest or what is less profitable for the broker. Similarly, and closer to therapy relationships, a lawyer cannot represent both plaintiff and defendant in a dispute - the lawyer cannot act for two principals and be loyal to both.

A “fiduciary is, in law, a person who is obliged to discharge faithfully a responsibility of trust toward another. Among the common fiduciary relationships are guardian to ward, parent to child, lawyer to client, corporate director to corporation, trustee to trust, and business partner to business partner.” 
The Columbia Electronic Encyclopedia Copyright © 2004.

And psychologist to client. A cornerstone of counseling is the moral principle of trust, the client’s belief that the clinician will be honest and tell the whole truth, will not conceal any important relevant information, and will maintain the confidentiality of what is entrusted to the clinician. Underlying all of these is the principle of fiduciary responsibility - to not take advantage of the client for the benefit of the clinician. Although it may not always be apparent, concealing some facts, inadequate documentation, or breaching confidentiality would only be done for some advantage to the clinician. The promise of non-exploitation provides the safety to reveal and explore thoughts, feelings, history, plans or anything else which would have to be concealed out of fear of the consequences to the client.

“In a fiduciary relation one person justifiably reposes confidence, good faith, reliance and trust in another whose aid, advice or protection is sought in some matter. In such a relation good conscience requires one to act at all times for the sole benefit and interests of another, with loyalty to those interests.” Wikipedia, accessed December 17, 2009.

Respect and autonomy

While such efforts are legally and ethically imposed they also can be empowering of the client and clinically useful in the therapy.

The fiduciary principle expands the principles of respect and trustworthiness in specific and helpfully guiding ways. It recognizes the multiple differences between the parties and extends that knowledge into ethical rules. The fiduciary duty requires one to explore, identify and recognize differences. It requires one to weigh and to adapt to these differences; to respect and dignify them. 

With respect for the other comes trust by the other in our loyalty to them. The enactment of such trust can be seen in at least  two relevant and related ethical responsibilities placed on clinicians: 
to ensure fully informed consent about all aspects of the therapeutic relationship both initially and as the relationship evolves and is extended, and 
to make every necessary effort to ensure and protect the privacy of the patient’s information.  

Multiple relationships

We have an ethical obligation to avoid harmful multiple relationships. But why? 

First, because each role has obligations and expectations and these may be in conflict. Multiple roles require choices of actions and the therapist might choose to meet his or her needs or wants before the needs of the client or just be seen by the client as putting his or her needs first. Simply put, a fiduciary may not and does not have a conflict of interests. 

A second element is that by foregoing any relationship other than client to therapist we absolutely prevent even accidental breaches of confidentiality. If we are not members of the client’s work setting or deal with relatives of the client we cannot even accidentally reveal information vouchsafed to us with the expectation of confidentiality (that we will not share it with another).

Among our responses can be a withdrawal from one role or the seeking of informed consent to allow limited role overlap. For example, we can inform about the ways a friendship would conflict with therapy and ask for consent to forego the friendship’s benefits. If we fail to resolve he conflicted interests we will breach our fiduciary duty and thus be open to both accusations of ethical failure and the possibilities of malpractice suits. 

In our work, preserving privacy, respecting differences, and supporting autonomy are aspects of our fiduciary duty. Carrying out this duty consistently and thoroughly - professionally - enables us to deserve the trust of those who seek our help.

Appreciation to Sam Knapp for his helpful comments.
December 17, 2009 and June 26, 2012.