AI-assisted euthanasia and the issue of autonomy

Tam-Tri Le
Phenikaa University (Hanoi, Vietnam)
https://orcid.org/0000-0003-3384-4827

November 4, 2022

The ability to choose is a direct expression of free will. However, as sophisticated social creatures, humans theoretically apply many behavioral restrictions on themselves to protect children and maintain a stable environment for co-existence. These restrictions are enforced by the law or established as social norms. For example, murder, arson, and robbery are crimes accompanied by severe legal punishment, while lesser forms of lying, verbal abuse, and disrespectfulness are collectively discouraged. In brief, many behaviors are technically viable but socially unacceptable to various degrees.

On this aspect, suicide is in the grey zone. Most countries in the world do not ban the act of suicide on a legal level. However, suicide is strongly discouraged and often actively prevented in almost all societies. The collective perception of suicide across cultures is relatively consistent. For example, all major religions in the world have condemned suicide throughout history [1]. But it is also obvious that suicidal intention is heavily individual, and the corresponding behaviors are up to oneself to decide [2]. Thus, one of the biggest issues regarding euthanasia is the notion of freedom of choice, besides ethical implications, legal complications, and risks of exploitation.

Euthanasia is assisted suicide (commonly with the help of physicians), mostly for patients with incurable illnesses to end their suffering quickly. In typical cases of autonomous euthanasia, the patients’ decisions may be deemed subjectively reasonable. However, human existence involves social interactions, and the death of patients also impacts their loved ones, the assisting physicians, the legal system, and public perception. For example, a study in Croatia found that the general public was largely against the idea of euthanasia [3]. While it is impossible to eliminate social connections (unless one lives in complete isolation), it is possible to increase one’s autonomy by lessening the linked impacts. One of the solutions in this direction is AI-assisted euthanasia.

Artificial intelligence (AI) can automate the assistance that patients require in euthanasia without the act of “killing” done by another human. Contemporary methods depend a lot on the moral considerations of the assisting physicians as well as their relationship with the patients [4]. Philip Nitschke – an advocator for euthanasia, often called “Dr. Death” – believes that a physician does not have to involve in assisted suicide. Thus he wants to create the tools necessary to help patients carry out their intended behavior [5]. It started rather simple, as he hooked up a laptop to the syringe to let his patients make the final decision on their own. Recently, he invented the Sarco pod to ease the procedure further and is now finalizing this product.


Figure: The Sarco pod (by Ratel, CC BY-SA 4.0)

The Sarco pod asks the user several short questions before activation by a four-digit code, serving as documents to be handed to the authority. When activated, the pod releases nitrogen gas to make the user pass out and quickly die by asphyxiation. Nitschke thinks that giving patients more autonomy in the decision of their deaths is to give them dignity in their final moments [5].

But beyond this simple automation, those who support euthanasia are also working on using AI to determine the mental capacity of patients demanding assisted suicide (instead of human psychiatrists). This may further reduce the subjective involvement of third parties. However, it is undoubtedly a highly debatable endeavor due to the conflicts among scientists, doctors, governments, and patients’ family members about its ethical implications.

When viewing a psychosocial phenomenon, it is important to look at it from the perspective of each involved party (here: the patient, the physician, the family members, etc.) and a neutral observer. Due to how information is evaluated in the human mind, problems involving subjective values such as emotions and ethics make it less likely to simulate an unbiased viewpoint upon examination [6]. But it should be noted that such subjective notions are crucial in human thinking and cannot be discarded in the hope of achieving absolute objectivity in a psychosocial investigation or discussion. In fact, incorporating intrinsic human subjectivity into the reasoning and analysis following a logical and rigorous protocol such as the Bayesian Mindsponge Framework (BMF) is quite advantageous in dealing with this kind of problem [6].

Regardless of supporting technologies, the core question in this problem will probably remain a very difficult one far into the future: Is a human supposed to have full control over the state of his/her own existence in the life-and-death duality?

References

[1] Vuong QH, Nguyen MH, Le TT. (2021). A Mindsponge-Based Investigation into the Psycho-Religious Mechanism Behind Suicide Attacks. De Gruyter.

[2] Nguyen MH, et al. (2021). Alice in Suicideland: Exploring the Suicidal Ideation Mechanism through the Sense of Connectedness and Help-Seeking Behaviors. International Journal of Environmental Research and Public Health, 18(7), 3681.

[3] Borovecki A, et al. (2022). Attitudes about withholding or withdrawing life-prolonging treatment, euthanasia, assisted suicide, and physician assisted suicide: A cross-sectional survey among the general public in Croatia. BMC Medical Ethics, 23(1), 13.

[4] Zwol M, Boer F, Evans N, Widdershoven G. (2022). Moral values of Dutch physicians in relation to requests for euthanasia: A qualitative study. BMC Medical Ethics, 23(1), 94.

[5] Heaven WD. (2022). The messy morality of letting AI make life-and-death decisions. MIT Technology Review.

[6] Vuong QH, La VP, Nguyen MH. (Eds.). (2022). The mindsponge and BMF analytics for innovative thinking in social sciences and humanities. De Gruyter.