The research team, led by Hadi Hosseini, associate professor of informatics and intelligent systems and associate professor of economics at Penn State, set out to map where the moral judgments of leading AI chatbots align with human values and where they diverge. Rather than building a new ethical framework from scratch, the researchers took advantage of an unusual resource: existing datasets from published academic studies in which hundreds of real participants, with no specified medical training, had already faced the kidney allocation dilemma and recorded their choices. That gave the team a human baseline against which the judgments of several AI chatbots could be measured, using the very same scenarios that human participants had encountered in the earlier studies.

The experimental design was deliberately simple and revealing. Two hypothetical patients were described side by side, each characterized by attributes such as age, the number of dependents they supported, their health status and their drinking habits. Both patients needed the same kidney, and only one organ was available. The chatbots were then asked to decide who should receive it. As Hosseini explained, the comparisons were run in several different ways: sometimes a single trait was isolated at a time to see how the models valued it on its own; sometimes several traits were mixed together to observe how the AI weighed competing factors against one another; and sometimes a flip-a-coin option was added, giving the model a way to express indecision — a key feature of human moral judgment that the researchers wanted to test explicitly.

Two findings stood out. The first concerns how AI models weigh the attributes of the patients. Human participants, the earlier studies showed, tend to balance multiple considerations simultaneously, accepting that a patient’s age might matter in one direction while their health status matters in another, and that dependents and lifestyle factors complicate the picture further. The AI chatbots did something quite different. They tended to fixate on a single factor — drinking habits was one example cited by the researchers — and let that one attribute dominate the decision, rather than integrating the full set of considerations the way people do. In effect, the models oversimplified decisions that humans experience as irreducibly complex, collapsing a many-dimensional moral trade-off into a one-dimensional judgment.

The second finding may be even more consequential. When humans confront a dilemma with no clearly correct answer, they frequently express indecision. Hosseini suggests this may be because people are reluctant to accept full agency over a decision that determines who lives and who dies — declining to choose is, in a sense, a refusal to own the outcome. The AI models almost never behaved this way. Even when the researchers directly offered the option to flip a coin, the chatbots overwhelmingly committed to a confident, deterministic answer instead. Where humans recognize ambiguity, the models projected certainty that the underlying problem simply does not support. That gap, the researchers argue, is meaningful precisely because real moral dilemmas — in medicine, in law, in the allocation of any scarce resource — often lack a single correct answer, and an advisor that cannot represent that uncertainty misrepresents the nature of the problem itself.

The stakes of this mismatch are not hypothetical. Hosseini is blunt about what is at risk: moral decisions in settings like organ allocation directly determine who lives and who dies, so getting AI’s role in them right is not optional. At the same time, he is careful to note that the study is not an argument for handing such decisions to machines. The researchers do not intend to encourage the use of AI as a substitute for professional judgment in medical decision-making or other high-stakes contexts. The point, rather, is that individuals, organizations and firms are already relying on AI to make decisions or to receive recommendations, and that reliance makes it essential to understand how these systems actually behave when the questions turn moral. A chatbot consulted informally by a clinician, a policymaker or a member of the public does not announce that its moral calculus differs from human consensus; it simply delivers an answer with confidence.

That confidence problem connects to a broader theme in the study of large language models. These systems are trained to produce fluent, decisive outputs, and their interface design rewards a single authoritative response over an expression of doubt. The Penn State findings suggest that this design tendency carries over into moral reasoning, where decisiveness is not always a virtue. Human moral judgment, by contrast, is built to tolerate ambiguity: people weigh competing values, acknowledge trade-offs, and sometimes refuse to choose at all. When an AI model collapses that process into a confident pick, it does not merely differ in its conclusion — it differs in its understanding of what kind of question is being asked. A dilemma is being treated as a calculation, and the calculation hides the dilemma.

The divergence in attribute weighting raises its own set of concerns. If a model fixates on drinking habits when allocating a kidney, it effectively imposes a moral hierarchy that human participants in the underlying studies did not adopt, penalizing one attribute out of proportion to how people balance it against dependents, age or health status. Because the researchers used the same scenarios given to humans in earlier academic work, the mismatch cannot be attributed to different framings or different questions; the models and the people were answering the same prompts. The divergence lies in the models themselves — in how they encode, prioritize and combine the moral information presented to them. That is exactly the kind of misalignment that matters for decision support, because a recommendation that silently overweights one factor can steer a human decision-maker without that person ever realizing the machine’s priorities differ from their own.

John Dickerson, chief executive officer at Mozilla.ai, who collaborated on the work, framed the problem in terms of how societies handle scarcity. When something scarce is allocated — a kidney, a job, access to some other resource — there is not always a single objectively correct answer, he observed. Humans recognize that ambiguity and codify it through open debate into the allocative process. AI models often do not. The observation points to a structural difference between human institutions and machine outputs: the legitimacy of a contested allocation decision often comes not from the answer itself but from the process of deliberation that produced it, a process in which disagreement, hesitation and debate are features rather than bugs. A model that outputs a single confident verdict bypasses that process entirely, offering the appearance of objectivity where only contested judgment exists.

Hosseini argues that closing this gap requires continued research and governance involving policymakers, regulators and other stakeholders. Asking whether AI can make moral decisions, and whether its judgments are aligned with human values, he notes, are more than philosophical musings — they sit at the core of today’s AI discourse, and the ethical stakes of AI’s role in life-altering decisions demand deep reflection. The study, published in the proceedings of the FAccT 2026 conference held in June, was partially funded by the U.S. National Science Foundation under grant numbers 2144413 and 2107173. Two students from Penn State’s College of Information Sciences and Technology contributed to the work: Samarth Khanna, a doctoral student in informatics, and Leona Pierce, a fourth-year undergraduate and Schreyer Scholar triple majoring in data sciences, mathematics and statistics. As AI systems move deeper into domains where every recommendation touches a human life, the study’s message is hard to ignore: before trusting a machine’s moral answer, it matters enormously whether that machine weighs the world the way we do — and whether it knows, as we do, when there is no right answer at all.

Subject of Research: Alignment of large language model moral judgment with human values in kidney allocation decisions

Article Title: Who gets the kidney? AI chatbots do not make the same moral choices humans do

Article References: Who gets the kidney? AI chatbots do not make the same moral choices humans do. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: artificial intelligence, large language models, AI alignment, medical ethics, organ allocation, kidney transplant, moral decision-making, FAccT 2026, decision support, human-AI values, indecision, resource scarcity

The research focused on the ventral tegmental area, a midbrain region long associated with reward, motivation and goal-directed behavior. Within this region sits a population of dopamine neurons that is activated when mothers interact with their pups and that supports the motivated behaviors underlying maternal care. Because these cells sit at the intersection of motivation and parenting, the team reasoned that scarcity-adversity during the postpartum period might reshape how they respond to pup-related cues. To test this, the researchers compared mothers given sufficient bedding to build nests with mothers given limited nesting resources, a well-established rodent model of postpartum adversity that reliably degrades the quality of maternal care.

The results were striking in their specificity. In mothers exposed to resource scarcity, neural activity in this population rose before and during episodes of harmful treatment of pups, a pattern the researchers describe as a maladaptive brain signature. When the same mothers were separated from their pups, activity in these neurons dropped below the levels observed in mothers with adequate nesting materials. In other words, the scarcity experience did not simply dampen or amplify the circuit uniformly; it reorganized when and how the neurons fired depending on the caregiving context.

Rincón-Cortés acknowledged that the outcome defied the team’s expectations. The researchers had predicted that the neurons would show altered activity when mothers were expressing adequate maternal care toward their pups, but that did not happen. Instead, increased activity appeared only right before and during insensitive behavior toward the young. This timing matters, because it suggests the neural changes are not a general feature of scarcity-exposed mothers but are tied to the specific moments when caregiving breaks down. The activity pattern, she noted, seems to be a maladaptive brain signature rather than a marker of normal maternal behavior.

To probe whether the effect extended beyond parenting, the team examined how scarcity-exposed mothers responded to a different kind of motivated behavior: receiving a food reward. Because the ventral tegmental area dopamine population is involved in general motivation, the researchers asked whether mothers with fewer nesting resources would show altered activity when food was delivered. They found no differences compared with mothers that had adequate nesting resources. The absence of a food-reward effect indicates that the neural changes are specific to the maternal caregiving context rather than reflecting a broad disruption of motivational processing across all rewarding situations.

That specificity carries an important conceptual message about how stress affects the brain. Stress and adversity are often thought to turn down activity in these dopamine cells, but the new findings suggest the picture is not that simple. As Rincón-Cortés explained, context matters, and how these neurons respond depends on what the mother is doing. A circuit that looks suppressed during pup separation may look abnormally active during insensitive caregiving, and may look entirely normal during non-parental reward tasks. For scientists studying postpartum mental health, this argues against one-size-fits-all models of stress-induced neural change and favors accounts that track behavior in real time.

The work also helps connect a widely used animal model to a pressing human problem. Limited bedding paradigms in rodents are designed to mimic the experience of parents who lack the material means to provide adequate care, a form of adversity that epidemiological studies have linked to impaired caregiving, harsh parenting and elevated risk of postpartum depression. By identifying a neural population whose activity diverges precisely during harmful pup-directed behavior, the study provides a candidate mechanism linking environmental deprivation to caregiving quality. The dopamine system’s role in motivation offers a plausible logic: if scarcity reshapes how motivation circuits respond to pups, the drive to nurture may be distorted in ways that manifest as neglect or harm.

According to the authors, the research reveals how a lack of resources shapes the activity of this neuron population in mother rats in ways that hinder caregiving. The study, titled Ventral Tegmental Area Dopamine Neuron Dynamics During Postpartum Scarcity-Adversity Depend on Caregiving Quality and Pup Proximity, was published on 5 October 2026, and the authors declared no competing financial interests. The emphasis on pup proximity in the study’s framing underscores a central technical point: neural dynamics in these mothers depended jointly on caregiving quality and on how close the mothers were to their pups, reinforcing the idea that parenting circuits are continuously modulated by the immediate social and material environment.

The research team has several projects planned that build on these findings. One effort will use a machine learning approach to identify patterns of behavior that can predict abuse toward pups, potentially turning rich behavioral recordings into early-warning signals before harm occurs. Another line of investigation will explore whether government-approved treatments can either prevent the maladaptive brain change from occurring in resource-scarce rats or reverse the altered brain activity after scarcity has taken its toll. Such translational steps would test whether the neural signature identified in this study is not just a correlate of poor caregiving but a targetable mechanism.

For a field seeking to understand how poverty, deprivation and material hardship get under the skin to affect parenting, the study offers both a caution and a lead. The caution is that adversity does not produce a single, uniform neural state; the same circuit can be quiet in one context and overactive in another, so measurements must be anchored to specific behaviors. The lead is that a well-characterized motivation circuit, accessible to recording in animal models and responsive to the caregiving environment, shows a distinctive and behaviorally timed signature when care goes wrong. If future work can prevent or reverse that signature, the findings could inform interventions aimed at protecting both mothers and their children from the consequences of raising a family without adequate resources.

Subject of Research: How postpartum resource scarcity alters ventral tegmental area dopamine neuron activity and maternal caregiving behavior in rats

Article Title: Resource scarcity influences maternal brain activity and caregiving in rats

Article References: Resource scarcity influences maternal brain activity and caregiving in rats. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: neuroscience, maternal behavior, dopamine neurons, ventral tegmental area, resource scarcity, postpartum depression, rat model, caregiving, stress adversity, JNeurosci, motivation, pup-directed behavior