Humans feel too special for machines to score their morals

Purcell, Z. A., & Jean‐François Bonnefon. (2023).

PNAS Nexus, 2(6).

https://doi.org/10.1093/pnasnexus/pgad179

Abstract

Artificial intelligence (AI) can be harnessed to create sophisticated social and moral scoring systems—enabling people and organizations to form judgments of others at scale. However, it also poses significant ethical challenges and is, subsequently, the subject of wide debate. As these technologies are developed and governing bodies face regulatory decisions, it is crucial that we understand the attraction or resistance that people have for AI moral scoring. Across four experiments, we show that the acceptability of moral scoring by AI is related to expectations about the quality of those scores, but that expectations about quality are compromised by people's tendency to see themselves as morally peculiar. We demonstrate that people overestimate the peculiarity of their moral profile, believe that AI will neglect this peculiarity, and resist for this reason the introduction of moral scoring by AI.‌

Significance Statement

The potential use of artificial intelligence (AI) to create sophisticated social and moral scoring systems poses significant ethical challenges. To inform the regulation of this technology, it is critical that we understand the attraction or resistance that people have for AI moral scoring. This project develops that understanding across four empirical studies—demonstrating that people overestimate the peculiarity of their moral profile, believe that AI will neglect this peculiarity, and resist for this reason the introduction of moral scoring by AI.

The link to the research is above.

My summary:

Here is another example of "myside bias" in which humans base decisions based on their uniqueness or better than average hypothesis.  This research study investigated whether people would accept AI moral scoring systems. The study found that people are unlikely to accept such systems, in large part because they feel too special for machines to score their personal morals.

Specifically, the results showed that people were more likely to accept AI moral scoring systems if they believed that the systems were accurate. However, even if people believed that the systems were accurate, they were still less likely to accept them if they believed that they were morally unique.

The study's authors suggest that these findings may be due to the fact that people have a strong need to feel unique and special. They also suggest that people may be hesitant to trust AI systems to accurately assess their moral character.

Key findings:

• People are unlikely to accept AI moral scoring systems, in large part because they feel too special for machines to score their personal morals.
• People's willingness to accept AI moral scoring is influenced by two factors: their perceived accuracy of the system and their belief that they are morally unique.
• People are more likely to accept AI moral scoring systems if they believe that the systems are accurate. However, even if people believe that the systems are accurate, they are still less likely to accept them if they believe that they are morally unique.

Minds of machines: The great AI consciousness conundrum

Grace Huckins

MIT Technology Review

Originally published 16 October 23

Here is an excerpt:

At the breakneck pace of AI development, however, things can shift suddenly. For his mathematically minded audience, Chalmers got concrete: the chances of developing any conscious AI in the next 10 years were, he estimated, above one in five.

Not many people dismissed his proposal as ridiculous, Chalmers says: “I mean, I’m sure some people had that reaction, but they weren’t the ones talking to me.” Instead, he spent the next several days in conversation after conversation with AI experts who took the possibilities he’d described very seriously. Some came to Chalmers effervescent with enthusiasm at the concept of conscious machines. Others, though, were horrified at what he had described. If an AI were conscious, they argued—if it could look out at the world from its own personal perspective, not simply processing inputs but also experiencing them—then, perhaps, it could suffer.

AI consciousness isn’t just a devilishly tricky intellectual puzzle; it’s a morally weighty problem with potentially dire consequences. Fail to identify a conscious AI, and you might unintentionally subjugate, or even torture, a being whose interests ought to matter. Mistake an unconscious AI for a conscious one, and you risk compromising human safety and happiness for the sake of an unthinking, unfeeling hunk of silicon and code. Both mistakes are easy to make. “Consciousness poses a unique challenge in our attempts to study it, because it’s hard to define,” says Liad Mudrik, a neuroscientist at Tel Aviv University who has researched consciousness since the early 2000s. “It’s inherently subjective.”

The info is here.

Here is my take.

There is an ongoing debate about whether artificial intelligence can ever become conscious or have subjective experiences like humans. Some argue AI will inevitably become conscious as it advances, while others think consciousness requires biological qualities that AI lacks.

Philosopher David Chalmers has proposed a "hard problem of consciousness" - explaining how physical processes in the brain give rise to subjective experience. This issue remains unresolved.

AI systems today show no signs of being conscious or having experiences. But some argue as AI becomes more sophisticated, we may need to consider whether it could develop some level of consciousness.

Approaches like deep learning and neural networks are fueling major advances in narrow AI, but this type of statistical pattern recognition does not seem sufficient to produce consciousness.

Questions remain about whether artificial consciousness is possible or how we could detect if an AI system were to become conscious. There are also ethical implications regarding the rights of conscious AI.

Overall there is much speculation but no consensus on whether artificial general intelligence could someday become conscious like humans are. The answer awaits theoretical and technological breakthroughs.

Private UK health data donated for medical research shared with insurance companies

Shanti Das

The Guardian

Originally poste 12 Nov 23

Sensitive health information donated for medical research by half a million UK citizens has been shared with insurance companies despite a pledge that it would not be.

An Observer investigation has found that UK Biobank opened up its vast biomedical database to insurance sector firms several times between 2020 and 2023. The data was provided to insurance consultancy and tech firms for projects to create digital tools that help insurers predict a person’s risk of getting a chronic disease. The findings have raised concerns among geneticists, data privacy experts and campaigners over vetting and ethical checks at Biobank.

Set up in 2006 to help researchers investigating diseases, the database contains millions of blood, saliva and urine samples, collected regularly from about 500,000 adult volunteers – along with medical records, scans, wearable device data and lifestyle information.

Approved researchers around the world can pay £3,000 to £9,000 to access records ranging from medical history and lifestyle information to whole genome sequencing data. The resulting research has yielded major medical discoveries and led to Biobank being considered a “jewel in the crown” of British science.

Biobank said it strictly guarded access to its data, only allowing access by bona fide researchers for health-related projects in the public interest. It said this included researchers of all stripes, whether employed by academic, charitable or commercial organisations – including insurance companies – and that “information about data sharing was clearly set out to participants at the point of recruitment and the initial assessment”.

The info is here.

Here is my summary:

Private health data donated by over half a million UK citizens for medical research has been shared with insurance companies, despite a pledge that it would not be used for this purpose. The data, which includes genetic information, medical diagnoses, and lifestyle factors, has been used to develop digital tools that help insurers predict a person's risk of getting a chronic disease. This raises concerns about the privacy and security of sensitive health data, as well as the potential for insurance companies to use the data to discriminate against people with certain health conditions.

How important is the end of humanity? Lay people prioritize extinction prevention but not above all other societal issues

Coleman, M. B., Caviola, L., et al.

(2023, October 21). 

https://doi.org/10.31234/osf.io/qn7k5

Abstract

Human extinction would mean the deaths of eight billion people and the end of humanity’s achievements, culture, and future potential. On several ethical views, extinction would be a terrible outcome. How do people think about human extinction? And how much do they prioritize preventing extinction over other societal issues? Across six empirical studies (N = 2,541; U.S. and China) we find that people consider extinction prevention a global priority and deserving of greatly increased societal resources. However, despite estimating the likelihood of human extinction to be 5% this century (U.S. median), people believe the odds would need to be around 30% for it to be the very highest priority. Consequently, people consider extinction prevention to be only one among several important societal issues. People’s judgments about the relative importance of extinction prevention appear relatively fixed and are hard to change by reason-based interventions.

The research is here.

Here is my take:

The study found that lay people rated extinction prevention as more important than addressing climate change, poverty, and inequality. However, they rated extinction prevention as less important than promoting peace and security, and ensuring the well-being of future generations.

The study's authors suggest that these findings may be due to the fact that lay people perceive extinction prevention as a more existential threat than other societal issues. They also suggest that lay people may be more likely to prioritize extinction prevention if they believe that it is achievable.

Key findings:

• Lay people prioritize extinction prevention, but not above all other societal issues.
• Lay people rated extinction prevention as more important than addressing climate change, poverty, and inequality.
• Lay people rated extinction prevention as less important than promoting peace and security, and ensuring the well-being of future generations.
• The study's authors suggest that these findings may be due to the fact that lay people perceive extinction prevention as a more existential threat than other societal issues.

Prosociality should be a public health priority

Kubzansky, L.D., Epel, E.S. & Davidson, R.J. 

Nat Hum Behav (2023).

https://doi.org/10.1038/s41562-023-01717-3

Standfirst:

Hopelessness and despair threaten health and longevity. We urgently need strategies to counteract these effects and improve population health. Prosociality contributes to better mental and physical health for individuals, and for the communities in which they live. We propose that prosociality should be a public health priority.

Comment:

The COVID-19 pandemic produced high levels of stress, loneliness, and mental health problems, magnifying global trends in health disparities.1 Hopelessness and despair are growing problems particularly in the U.S. The sharp increase in rates of poor mental health is problematic in its own right, but poor mental health also contributes to greater morbidity and mortality. Without action, we will see steep declines in global population health and related costs to society. An approach that is “more of the same” is insufficient to stem the cascading effects of emotional ill-being. Something new is desperately needed.

To this point, recent work called on the discipline of psychiatry to contribute more meaningfully to the deaths of despair framework (i.e., conceptualizing rises in suicide, drug poisoning and alcoholic liver disease as due to misery of difficult social and economic circumstances).2 Recognizing that simply expanding mental health services cannot address the problem, the authors noted the importance of population-level prevention and targeting macro-level causes for intervention. This requires identifying upstream factors causally related to these deaths. However, factors explaining population health trends are poorly delineated and focus on risks and deficits (e.g., adverse childhood experiences, unemployment). A ‘deficit-based’ approach has limits as the absence of a risk factor does not inevitably indicate presence of a protective asset; we also need an ‘assetbased’ approach to understanding more comprehensively the forces that shape good health and buffer harmful effects of stress and adversity.

A pdf is here.

My take:

Prosociality refers to positive behaviors and beliefs that benefit others. It is a broad concept that encompasses many different qualities, such as altruism, trust, reciprocity, compassion, and empathy.

Research has shown that prosociality has a number of benefits for both individuals and communities. For individuals, prosociality can lead to improved mental and physical health, greater life satisfaction, and stronger social relationships. For communities, prosociality can lead to increased trust and cooperation, reduced crime rates, and improved overall well-being.

The authors of the article argue that prosociality should be a public health priority. They point out that prosociality can help to address a number of major public health challenges, such as loneliness, social isolation, and mental illness. They also argue that prosociality can help to build stronger communities and create a more just and equitable society.

Ignorance by Choice: A Meta-Analytic Review of the Underlying Motives of Willful Ignorance and Its Consequences

Vu, L., Soraperra, I., Leib, M., et al. (2023).

Psychological Bulletin, 149(9-10), 611–635.

https://doi.org/10.1037/bul0000398

Abstract

People sometimes avoid information about the impact of their actions as an excuse to be selfish. Such “willful ignorance” reduces altruistic behavior and has detrimental effects in many consumer and organizational contexts. We report the first meta-analysis on willful ignorance, testing the robustness of its impact on altruistic behavior and examining its underlying motives. We analyze 33,603 decisions made by 6,531 participants in 56 different treatment effects, all employing variations of an experimental paradigm assessing willful ignorance. Meta-analytic results reveal that 40% of participants avoid easily obtainable information about the consequences of their actions on others, leading to a 15.6-percentage point decrease in altruistic behavior compared to when information is provided. We discuss the motives behind willful ignorance and provide evidence consistent with excuse-seeking behaviors to maintain a positive self-image. We investigate the moderators of willful ignorance and address the theoretical, methodological, and practical implications of our findings on who engages in willful ignorance, as well as when and why.

Public Significance Statement

We present the first meta-analysis on willful ignorance—when individuals avoid information about the negative consequences of their actions to maximize personal outcomes—covering 33,603 decisions made by 6,531 participants across 56 treatment effects. Results demonstrate that the ability to avoid such information decreases altruistic behavior, and that seemingly altruistic behavior may not reflect a true concern for others.

The research is here.

Key findings of the meta-analysis include:

Prevalence of Willful Ignorance: Approximately 40% of participants in the analyzed studies chose to avoid learning about the negative impact of their actions on others.

Impact on Altruism: Willful ignorance significantly reduces altruistic behavior. When provided with information about the consequences of their actions, participants were 15.6 percentage points more likely to engage in altruistic acts compared to those who chose to remain ignorant.

Motives for Willful Ignorance: The study suggests that willful ignorance may serve as a self-protective mechanism to maintain a positive self-image. By avoiding information about the harm caused by their actions, individuals can protect their self-perception as moral and ethical beings.

One of the top concerns is moral decline of today’s youth, survey

Valerie Pritchett

27ABC News

Originally published 9 NOV 23

Yes, I do TV interviews as well.  The video plays after the commercial.

Discordant benevolence: How and why people help others in the face of conflicting values.

Cowan, S. K., Bruce, T. C., et al. (2022).

Science Advances, 8(7).

https://doi.org/10.1126/sciadv.abj5851

Abstract

What happens when a request for help from friends or family members invokes conflicting values? In answering this question, we integrate and extend two literatures: support provision within social networks and moral decision-making. We examine the willingness of Americans who deem abortion immoral to help a close friend or family member seeking one. Using data from the General Social Survey and 74 in-depth interviews from the National Abortion Attitudes Study, we find that a substantial minority of Americans morally opposed to abortion would enact what we call discordant benevolence: providing help when doing so conflicts with personal values. People negotiate discordant benevolence by discriminating among types of help and by exercising commiseration, exemption, or discretion. This endeavor reveals both how personal values affect social support processes and how the nature of interaction shapes outcomes of moral decision-making.

Here is my summary:

Using data from the General Social Survey and 74 in-depth interviews from the National Abortion Attitudes Study, the authors find that a substantial minority of Americans morally opposed to abortion would enact discordant benevolence. They also find that people negotiate discordant benevolence by discriminating among types of help and by exercising commiseration, exemption, or discretion.

Commiseration involves understanding and sharing the other person's perspective, even if one does not agree with it. Exemption involves excusing oneself from helping, perhaps by claiming ignorance or lack of resources. Discretion involves helping in a way that minimizes the conflict with one's own values, such as by providing emotional support or practical assistance but not financial assistance.

The authors argue that discordant benevolence is a complex phenomenon that reflects the interplay of personal values, social relationships, and moral decision-making. They conclude that discordant benevolence is a significant form of social support, even in cases where it is motivated by conflicting values.

In other words, the research suggests that people are willing to help others in need, even if it means violating their own personal values. This is because people also value social relationships and helping others. They may do this by discriminating among types of help or by exercising commiseration, exemption, or discretion.

Attitudes in an interpersonal context: Psychological safety as a route to attitude change

Itzchakov, G., & DeMarree, K. G. (2022).

Frontiers in Psychology, 13.

https://doi.org/10.3389/fpsyg.2022.932413

Abstract

Interpersonal contexts can be complex because they can involve two or more people who are interdependent, each of whom is pursuing both individual and shared goals. Interactions consist of individual and joint behaviors that evolve dynamically over time. Interactions are likely to affect people’s attitudes because the interpersonal context gives conversation partners a great deal of opportunity to intentionally or unintentionally influence each other. However, despite the importance of attitudes and attitude change in interpersonal interactions, this topic remains understudied. To shed light on the importance of this topic. We briefly review the features of interpersonal contexts and build a case that understanding people’s sense of psychological safety is key to understanding interpersonal influences on people’s attitudes. Specifically, feeling psychologically safe can make individuals more open-minded, increase reflective introspection, and decrease defensive processing. Psychological safety impacts how individuals think, make sense of their social world, and process attitude-relevant information. These processes can result in attitude change, even without any attempt at persuasion. We review the literature on interpersonal threats, receiving psychological safety, providing psychological safety, and interpersonal dynamics. We then detail the shortcomings of current approaches, highlight unanswered questions, and suggest avenues for future research that can contribute in developing this field.

The info is here.

This is part of the reason psychotherapy works.

My summary:

Attitudes are evaluations of people, objects, or ideas, and they can be influenced by a variety of factors, including interpersonal interactions. Psychological safety is a climate in which individuals feel safe to take risks, make mistakes, and be vulnerable. When people feel psychologically safe, they are more likely to express their true thoughts and feelings, which can lead to attitude change.

There are a number of ways that psychological safety can promote attitude change. First, feeling psychologically safe can make people more open-minded. When people feel safe, they are more likely to consider new information and perspectives, even if they challenge their existing beliefs. Second, psychological safety can increase reflective introspection. When people feel safe to be vulnerable, they are more likely to reflect on their own thoughts and feelings, which can lead to deeper insights and changes in attitude. Third, psychological safety can decrease defensive processing. When people feel safe, they are less likely to feel threatened by new information or perspectives, which makes them more open to considering them.

Research has shown that psychological safety can lead to attitude change in a variety of interpersonal contexts, including romantic relationships, friendships, and work teams. For example, one study found that couples who felt psychologically safe in their relationships were more likely to change their attitudes towards each other over time. Another study found that employees who felt psychologically safe in their teams were more likely to change their attitudes towards diversity and inclusion.