A Normative Approach to Artificial Moral Agency

Behdadi, D., Munthe, C.

Minds & Machines 30, 195–218 (2020).

https://doi.org/10.1007/s11023-020-09525-8

Abstract

This paper proposes a methodological redirection of the philosophical debate on artificial moral agency (AMA) in view of increasingly pressing practical needs due to technological development. This “normative approach” suggests abandoning theoretical discussions about what conditions may hold for moral agency and to what extent these may be met by artificial entities such as AI systems and robots. Instead, the debate should focus on how and to what extent such entities should be included in human practices normally assuming moral agency and responsibility of participants. The proposal is backed up by an analysis of the AMA debate, which is found to be overly caught in the opposition between so-called standard and functionalist conceptions of moral agency, conceptually confused and practically inert. Additionally, we outline some main themes of research in need of attention in light of the suggested normative approach to AMA.

Free will and Autonomy

Several AMA debaters have claimed that free will is necessary for being a moral agent (Himma 2009; Hellström 2012; Friedman and Kahn 1992). Others make a similar (and perhaps related) claim that autonomy is necessary (Lin et al. 2008; Schulzke 2013). In the AMA debate, some argue that artificial entities can never have free will (Bringsjord 1992; Shen 2011; Bringsjord 2007) while others, like James Moor (2006, 2009), are open to the possibility that future machines might acquire free will.Footnote15 Others (Powers 2006; Tonkens 2009) have proposed that the plausibility of a free will condition on moral agency may vary depending on what type of normative ethical theory is assumed, but they have not developed this idea further.

Despite appealing to the concept of free will, this portion of the AMA debate does not engage with key problems in the free will literature, such as the debate about compatibilism and incompatibilism (O’Connor 2016). Those in the AMA debate assume the existence of free will among humans, and ask whether artificial entities can satisfy a source control condition (McKenna et al. 2015). That is, the question is whether or not such entities can be the origins of their actions in a way that allows them to control what they do in the sense assumed of human moral agents.

An exception to this framing of the free will topic in the AMA debate occurs when Johnson writes that ‘… the non-deterministic character of human behavior makes it somewhat mysterious, but it is only because of this mysterious, non-deterministic aspect of moral agency that morality and accountability are coherent’ (Johnson 2006 p. 200). This is a line of reasoning that seems to assume an incompatibilist and libertarian sense of free will, assuming both that it is needed for moral agency and that humans do possess it. This, of course, makes the notion of human moral agents vulnerable to standard objections in the general free will debate (Shaw et al. 2019). Additionally, we note that Johnson’s idea about the presence of a ‘mysterious aspect’ of human moral agents might allow for AMA in the same way as Dreyfus and Hubert’s reference to the subconscious: artificial entities may be built to incorporate this aspect.

The question of sourcehood in the AMA debate connects to the independence argument: For instance, when it is claimed that machines are created for a purpose and therefore are nothing more than advanced tools (Powers 2006; Bryson 2010; Gladden 2016) or prosthetics (Johnson and Miller 2008), this is thought to imply that machines can never be the true or genuine source of their own actions. This argument questions whether the independence required for moral agency (by both functionalists and standardists) can be found in a machine. If a machine’s repertoire of behaviors and responses is the result of elaborate design then it is not independent, the argument goes. Floridi and Sanders question this proposal by referring to the complexity of ‘human programming’, such as genes and arranged environmental factors (e.g. education). 

The info is here.

Preventing an AI-related catastrophe

Benjamin Hilton

8000 Hours

Originally Published August 25th, 2022

Summary

We expect that there will be substantial progress in AI in the next few decades, potentially even to the point where machines come to outperform humans in many, if not all, tasks. This could have enormous benefits, helping to solve currently intractable global problems, but could also pose severe risks. These risks could arise accidentally (for example, if we don’t find technical solutions to concerns about the safety of AI systems), or deliberately (for example, if AI systems worsen geopolitical conflict). We think more work needs to be done to reduce these risks.

Some of these risks from advanced AI could be existential — meaning they could cause human extinction, or an equally permanent and severe disempowerment of humanity. There have not yet been any satisfying answers to concerns — discussed below — about how this rapidly approaching, transformative technology can be safely developed and integrated into our society. Finding answers to these concerns is very neglected, and may well be tractable. We estimate that there are around 300 people worldwide working directly on this.3 As a result, the possibility of AI-related catastrophe may be the world’s most pressing problem — and the best thing to work on for those who are well-placed to contribute.

Promising options for working on this problem include technical research on how to create safe AI systems, strategy research into the particular risks AI might pose, and policy research into ways in which companies and governments could mitigate these risks. If worthwhile policies are developed, we’ll need people to put them in place and implement them. There are also many opportunities to have a big impact in a variety of complementary roles, such as operations management, journalism, earning to give, and more — some of which we list below.

(cut)

When can we expect transformative AI?

It’s difficult to predict exactly when we will develop AI that we expect to be hugely transformative for society (for better or for worse) — for example, by automating all human work or drastically changing the structure of society. But here we’ll go through a few approaches.

One option is to survey experts. Data from the 2019 survey of 300 AI experts implies that there is 20% probability of human-level machine intelligence (which would plausibly be transformative in this sense) by 2036, 50% probability by 2060, and 85% by 2100. There are a lot of reasons to be suspicious of these estimates,8 but we take it as one data point.

Ajeya Cotra (a researcher at Open Philanthropy) attempted to forecast transformative AI by comparing modern deep learning to the human brain. Deep learning involves using a huge amount of compute to train a model, before that model is able to perform some task. There’s also a relationship between the amount of compute used to train a model and the amount used by the model when it’s run. And — if the scaling hypothesis is true — we should expect the performance of a model to predictably improve as the computational power used increases. So Cotra used a variety of approaches (including, for example, estimating how much compute the human brain uses on a variety of tasks) to estimate how much compute might be needed to train a model that, when run, could carry out the hardest tasks humans can do. She then estimated when using that much compute would be affordable.

Cotra’s 2022 update on her report’s conclusions estimates that there is a 35% probability of transformative AI by 2036, 50% by 2040, and 60% by 2050 — noting that these guesses are not stable.22

Tom Davidson (also a researcher at Open Philanthropy) wrote a report to complement Cotra’s work. He attempted to figure out when we might expect to see transformative AI based only on looking at various types of research that transformative AI might be like (e.g. developing technology that’s the ultimate goal of a STEM field, or proving difficult mathematical conjectures), and how long it’s taken for each of these kinds of research to be completed in the past, given some quantity of research funding and effort.

Davidson’s report estimates that, solely on this information, you’d think that there was an 8% chance of transformative AI by 2036, 13% by 2060, and 20% by 2100. However, Davidson doesn’t consider the actual ways in which AI has progressed since research started in the 1950s, and notes that it seems likely that the amount of effort we put into AI research will increase as AI becomes increasingly relevant to our economy. As a result, Davidson expects these numbers to be underestimates.

The info is here.

Have American jails become the inferior replacement for mental hospitals?

Matthew Rozsa

Salon.com

Originally posted 5 SEPT 22

Here is an excerpt:

"We've known for some time that this country's chief response to serious mental illness is incarceration, a fact that stands out because prisons are so clearly unsuited to treating mental illness," Wanda Bertram, Communications Strategist at Prison Policy Initiative, told Salon by email. "Our organization recently found that even though 43% of people in state prisons have been diagnosed with a mental disorder, only 26% have received some form of mental health treatment, and only 6% are currently receiving treatment."

Bertram added, "The readiness with which our justice system fast-tracks people with mental illnesses into prison, despite knowing that jail and prison settings won't make that person any better, speaks volumes about the system's ability to deliver justice."

Dr. Craig Haney, a psychologist who has studied the psychological effects of incarceration for decades and a psychology professor at the University of California, Santa Cruz, offered some insight into why America tends to incarcerate rather than help people with mental illnesses.

"The structural origins start with the history of two simultaneous trends that began in the early 1970s," Haney wrote to Salon. The first was the widespread closing of publicly-funded mental hospitals "in part on the promise that they would be replaced by more humane community-based treatment, a promise that was never kept" and the second was "the beginning of a decades-long 'tough-on-crime' era in which politicians competed with each other on who could criminalize the most things and impose the longest sentences. So we shrunk our mental health system and increased the size of our prison system."

Bertram also attributed the trend to imprison people who are mentally ill to ideological choices.

"I think the major problem is an ideology that says that if you have some kind of illness, including mental illness, you ought to be the primary person responsible for your own care," Bertram explained. "That's the ideology that props up our healthcare system, where sick people bear extraordinary costs and crushing debts. And it keeps us from asking why mental health services like therapy, psychiatry, and long-term care are not only expensive, but difficult to access." Pointing out that their report revealed roughly half of people in state prisons lacked any kind of health insurance prior to their arrest, Bertram concluded that "we continue to send people with mental disorders to prison, because there seems to be nowhere else for them to go."

The info is here.

Defining Their Own Ethics, Online Creators Are De Facto Therapists for Millions—Explosive Demand & Few Safeguards

Tantum Hunter

The Washington Post

Originally posted 29 AUG 22

Here are two excerpts:

In real life, mental health information and care are sparse. In the United States, 1 in 3 counties do not have a single licensed psychologist, according to the American Psychological Association, and Americans say cost is a top barrier to seeking mental health help. On the internet, however, mental health tips are everywhere: TikTok videos with #mentalhealth in the caption have earned more than 43.9 billion views, according to the analytics company Sprout Social, and mentions of mental health on social media are increasing year by year.

The growing popularity of the subject means that creators of mental health content are filling a health-care gap. But social media apps are not designed to prioritize accurate, helpful information, critics say, just whatever content draws the biggest reaction. Young people could see their deepest struggles become fodder for advertisers and self-promoters. With no road map even for licensed professionals, mental health creators are defining their own ethics.

“I don’t want to give anyone the wrong advice,” Moloney says. “I’ve met some [followers] who’ve just started crying and saying ‘thank you’ and stuff like that. Even though it seems small, to someone else, it can have a really big impact.”

As rates of depression and anxiety spiked during the pandemic and options for accessible care dwindled, creators shared an array of content including first-person accounts of life with mental illness and videos listing symptoms of bipolar disorder. In many cases, their follower counts ballooned.

(cut)

Ideally, social media apps should be one item in a collection of mental health resources, said Jodi Miller, a researcher at Johns Hopkins University School of Education who studies the relationships among young people, technology and stress.

“Young people need evidence-based sources of information outside the internet, from parents and schools,” Miller said.

Often, those resources are unavailable. So it’s up to consumers to decide what mental health advice they put stock in, Fisher-Quann said. For her, condescending health-care providers and the warped incentives of social media platforms haven’t made that easy. But she thinks she can get better — and that her followers can, too.

“It all has to come from a place of self-awareness and desire to get better. Communities can be extremely helpful for that, but they can also be extremely harmful for that,” she said.

The article is here.

Making psychiatry moral again: the role of psychiatry in patient moral development

McConnell D, Broome M, Savulescu, J.

Journal of Medical Ethics 

Published Online First: 19 August 2022.

doi: 10.1136/jme-2022-108442

Abstract

Psychiatric involvement in patient morality is controversial. If psychiatrists are tasked with shaping patient morality, the coercive potential of psychiatry is increased, treatment may be unfairly administered on the basis of patients’ moral beliefs rather than medical need, moral disputes could damage the therapeutic relationship and, in any case, we are often uncertain or conflicted about what is morally right. Yet, there is also a strong case for the view that psychiatry often works through improving patient morality and, therefore, should aim to do so. Our goal is to offer a practical and ethical path through this conflict. We argue that the default psychiatric approach to patient morality should be procedural, whereby patients are helped to express their own moral beliefs. Such a procedural approach avoids the brunt of objections to psychiatric involvement in patient morality. However, in a small subset of cases where patients’ moral beliefs are sufficiently distorted or underdeveloped, we claim that psychiatrists should move to a substantive approach and shape the content of those beliefs when they are relevant to psychiatric outcomes. The substantive approach is prone to the above objections but we argue it is nevertheless justified in this subset of cases.

(cut)

Helping people elaborate a conception of morality has little risk of coercion and damage to the therapeutic relationship because the patient is requesting that content and is not committed to a conflicting moral position. Of course, it would be wrong to simply indoctrinate the patient so, to avoid that, the process of moral development should be patient-led as much as possible. However, if the moral views the patient gravitates towards in this process are clearly unreasonable, then the psychiatrist has an obligation to guide the patient’s views back within the bounds of reasonableness.

Psychiatrists are well placed to affect substantive moral growth. Their skill for helping people understand and elaborate their subjective worlds can reveal where the moral growth required to treat mental illness and support flourishing might be most easily achieved. We suggest a pluralist approach where the psychiatrist draws on any moral reasons, arguments or insights that help the patient achieve moral growth. In order to tailor moral reasons to the patient, psychiatrists would benefit not only from training in normative moral theories (eg, contractualism, deontology, consequentialism) but also from familiarity with a diverse range of autobiographical or fictional narratives that illustrate how different moral views are experienced and put into practice. The latter would also provide material that the psychiatrist could draw on to help the patient develop moral aspects of their own self-narrative. In the near future, the substantive approach could also benefit from pharmacotherapies, such as psychedelics, which might help patients who consent to such treatment become more receptive to new moral reasons, beliefs and emotions.

The info is here.

A Systematic Review of Black People Coping With Racism: Approaches, Analysis, and Empowerment

Jacob, G., Faber, S. C., et al. (2022).

Perspectives on Psychological Science.

https://doi.org/10.1177/17456916221100509

Abstract

This article reviews the current research literature concerning Black people in Western societies to better understand how they regulate their emotions when coping with racism, which coping strategies they use, and which strategies are functional for well-being. A systematic review of the literature was conducted, and 26 studies were identified on the basis of a comprehensive search of multiple databases and reference sections of relevant articles. Studies were quantitative and qualitative, and all articles located were from the United States or Canada. Findings demonstrate that Black people tend to cope with racism through social support (friends, family, support groups), religion (prayer, church, spirituality), avoidance (attempting to avoid stressors), and problem-focused coping (confronting the situation directly). Findings suggest gender differences in coping strategies. We also explore the relationship between coping with physical versus emotional pain and contrast functional versus dysfunctional coping approaches, underscoring the importance of encouraging personal empowerment to promote psychological well-being. Findings may help inform mental-health interventions. Limitations include the high number of American-based samples and exclusion of other Black ethnic and national groups, which is an important area for further exploration.

From the Discussion section

Clinical implications

For clinicians seeking ways to support Black clients with racial trauma, the successful coping strategies enumerated here can serve as model starting points and should provide clients with greater agency and better outcomes (Heard-Garris et al., 2021; Hope et al., 2018) than the use of an ambiguous strategy. Therapy should be palpable positive affirmation; clients should feel validated and empowered. If they are religious, finding purpose in their experience even if it was negative can have a positive therapeutic effect. Helping clients find a coping strategy that affirms their intrinsic worth and beauty can also be profoundly therapeutic. If clients do not have affirmative social-support networks, or have dysfunctional social support, helping them find positively affirming support can be highly beneficial. Encouraging clients to create and make art, music, or prose out of their racist experience through positive reframing can be a transformative and proactive coping mechanism (Miller et al., 2020; Stuckey & Nobel, 2010). Certain forms of activism furthermore seem to have specific mental-health benefits (Heard-Garris et al., 2021; Montagno & Garrett-Walker, 2022; Riley et al., 2021). Ensuring that the coping mechanism chosen allows clients to reclaim their identity and dignity is essential. It is important to keep in mind that activism comes in many forms and may or may not involve formal protests or a Black Lives Matter event (E. K. Griffin & Armstead, 2020). Black clients can look for opportunities to promote antiracist change in their personal environments as well (work, school, community) through any number of prosocial means. For a cognitive-behavioral approach to helping clients with racial stress and trauma, see Williams et al. (in press).

The research is here.

Individualist moral principles and the expansion of the moral circle

Singh, M., & Hoffman, M. (2022, August 20).

https://doi.org/10.31235/osf.io/pqem7

Abstract

The last three centuries have witnessed a moral and political transformation. Groups previously denied equivalent moral standing—including propertyless men, women, ethnic and religious minorities, homosexuals, and slaves—became moral equals deserving of similar legal treatment. Here we argue that this process was driven by the reputational benefits of demonstrating commitment to individualist moral principles. These principles flourished in “fluid” social ecologies with high relational mobility and weak kinship institutions, both as individuals aimed to signal impersonal prosociality and as they strove to be governed under institutions that protected substitutable individuals unbound by formal obligations. As long as parties benefited from appearing committed to these principles, and denying rights to certain groups appeared inconsistent with these principles, then those parties were incentivized to grant those rights. Given the universalist nature of these principles, people signaling commitment were also incentivized to sanction rights-based violations in other countries, helping expand rights beyond their original context. We use this account to explain both expansions and contractions of the moral circle and reconcile the roles of ideas, markets, reasoning, reputation, the Catholic Church, argumentation, moral intuition, social organization, individual strategizing, and large-scale cultural evolution.

From the General Discussion

The first, which we’ll call sincere ideology, acknowledges that expansions occurred because of internalized principles yet rejects or downplays the role of trust. By this account, people have pushed for better treatment because of principles adopted for many potential reasons, such as argumentation or other socialization processes, yet desires to attract trust have been minimally involved.  For the sake of convenience, we refer this account as sincere ideology, although a better name might be sincere ideology without trust, given that our account permits that individuals will internalize and genuinely feel committed to principles to the extent that doing so helps cultivate trust.

The second alternative we call bargaining-for-rights. By this account, the moral circle has expanded because minority groups were better able to bargain for better treatment. An example of this account is Acemoglu and Robinson’s (2000) proposal that Western elites expanded the franchise as concessions to deter social unrest.

Both alternatives are consistent with some of the observations we and other researchers have reported. For instance, the sincere ideology account similarly predicts that revolutionaries who claim certain moral principles will create corresponding institutions after gaining power (sect. 4.2).  Bargaining-for-rights, meanwhile, is consistent with rights expanding with increasingly fluid social ecologies.  Yet on their own, each has difficulty explaining key patterns.

The research is here.

Permitting immoral behaviour: A generalized compensation belief hypothesis

Wang, X., Chen, Z., et al. (2022). 

British Journal of Psychology.

https://doi.org/10.1111/bjop.12593

Abstract

When are we more likely to permit immoral behaviours? The current research examined a generalized compensation belief hypothesis that individuals, as observers, would morally tolerate and accept someone paying forward unfair treatment to an innocent person as a means to compensate for the perpetrator's previously experienced mistreatment. Across five experiments (N = 1107) based on economic games (Studies 1–4) and diverse real-life scenarios (Study 5), we showed that participants, as observing third parties, were more likely to morally permit and engage in the same negative act once they knew about previous maltreatment of the perpetrator. This belief occurred even when the content of received and paid-forward maltreatment was non-identical (Study 2), when the negative treatment was received from a non-human target (Study 3) and when the maltreatment was intangible (e.g. material loss) or relational (e.g. social exclusion; Study 5). Perceived required compensation mediated the effect of previous maltreatment on moral permission (Studies 4 and 5). The results consistently suggest that people's moral permission of immoral behaviours is influenced by perpetrator's previous mistreatment, contributing to a better understanding of the nature and nuances of our sense of fairness and contextualized moral judgement.

From the General Discussion

The existence of a generalized compensation belief is in line with the theories of person-specific equity and equity with the world, which state that individuals attempt to maintain net equity (i.e. a balance between loss and gain) both within and across relationships (Austin & Walster, 1974; Homans, 1961).  Whereas traditional person-specific equity and equity with the world theories are mainly concerned about how I (i.e. the first-person perspective) actually form balanced relations with others (out of potential motives, e.g. self-interest and negative affect), generalized compensation belief examines individuals' perception of how general social interactions involving mistreatments should be, and it shows that individuals have a fundamental moral belief in balancing social interactions and permit paying-forward moral mistreatment, even when their self-interest remains relatively irrelevant.

At first sight, the generalized compensation belief may be inconsistent with the predictions of some theories. For example, the theory of morality-as-cooperation argues that solutions to problems of cooperation constitute human morality (Curry et al., 2019a, 2019b), thus whether an act is considered morally acceptable or not should depend on whether it promotes cooperation. Generalized compensation belief permits paying forward a harmful act and does not promote immediate interpersonal cooperation (if not promote the opposite). However, it is worth pointing out that the theory of morality-as-cooperation makes no specific predictions about how we make a moral judgement in particular proximate contexts.  The generalized compensation belief could still be compatible with the ultimate function of ensuring generalized reciprocity and promoting cooperation in the end, a possibility worth examining using evolutionary psychology approaches.

The research is here.

COVID-19 and Politically Motivated Reasoning

Maguire, A., Persson, E., Västfjäll, D., & 

Tinghög, G. (2022). Medical Decision Making.

https://doi.org/10.1177/0272989X221118078

Abstract

Background

During the COVID-19 pandemic, the world witnessed a partisan segregation of beliefs toward the global health crisis and its management. Politically motivated reasoning, the tendency to interpret information in accordance with individual motives to protect valued beliefs rather than objectively considering the facts, could represent a key process involved in the polarization of attitudes. The objective of this study was to explore politically motivated reasoning when participants assess information regarding COVID-19.

Design

We carried out a preregistered online experiment using a diverse sample (N = 1500) from the United States. Both Republicans and Democrats assessed the same COVID-19–related information about the health effects of lockdowns, social distancing, vaccination, hydroxychloroquine, and wearing face masks.

Results

At odds with our prestated hypothesis, we found no evidence in line with politically motivated reasoning when interpreting numerical information about COVID-19. Moreover, we found no evidence supporting the idea that numeric ability or cognitive sophistication bolster politically motivated reasoning in the case of COVID-19. Instead, our findings suggest that participants base their assessment on prior beliefs of the matter.

Conclusions

Our findings suggest that politically polarized attitudes toward COVID-19 are more likely to be driven by lack of reasoning than politically motivated reasoning—a finding that opens potential avenues for combating political polarization about important health care topics.

Highlights

• Participants assessed numerical information regarding the effect of different COVID-19 policies.

• We found no evidence in line with politically motivated reasoning when interpreting numerical information about COVID-19.

• Participants tend to base their assessment of COVID-19–related facts on prior beliefs of the matter.

• Politically polarized attitudes toward COVID-19 are more a result of lack of thinking than partisanship.

The research is here.