Grief is a universal human experience. The experience of grief, though described as valuable and transformative, is also painful. As a response to intense grief experienced by bereaved individuals, case studies of individuals using artificial intelligence (AI) software to communicate with simulations of deceased loved ones have been noted. However, it is unclear if such an approach promotes the bereaved’s wellbeing. Considering the maturation of literature of ethical issues in AI for grief support, this scoping review aims to evaluate key ethical considerations to enhance subsequent debate. A scoping review of relevant articles published between 1 January 2000 and 26 October 2025 in 10 databases across medicine, philosophy, social sciences, and computer sciences was carried out. Thematic analysis of relevant articles was performed. 2387 non-duplicate abstracts were screened, 121 full texts reviewed and 36 papers were included. Themes identified were notions of grief, ethical considerations surrounding the use of AI for grief support, and varied recommendations for design and implementation. Through this review, it was noted that most articles were sensitive to theoretical notions of grief and focused on ethical considerations surrounding the impact of AI use on grieving processes for the bereaved and respect for the deceased. Further ethical work can provide boundaries for the use of AI for grief support which should also be sensitive to individual, contextual, and sociocultural dimensions of grief.
Cognitive enhancement involves using substances or technologies to improve mental performance in healthy individuals. While methods such as caffeine are widely accepted, others, including prescription stimulants and non-invasive brain stimulation (tDCS), evoke ethical concern. This study examined how university students evaluate three forms of cognitive enhancement (natural, pharmacological, brain stimulation) across five moral domains: academic fairness, free will, authenticity/naturalness, self-identity, and safety. We also tested how evaluations were shaped by framing (preservation vs. enhancement) and priming (self-affecting vs. non self-affecting). A total of 449 undergraduate psychology students (81.7
The rapid advance of neurotechnology has produced a vibrant scholarly and legislative debate around neurorights — a proposed generation of human rights designed to protect mental privacy, cognitive liberty, mental integrity, and psychological continuity against the most threatening applications of neurotechnology. This article argues that the neurorights framework, while representing a genuine and important advance in the governance of neurotechnology, contains a structural blind spot: it was designed to protect individuals from non-consensual external interferences and is not well equipped to address scenarios in which individuals voluntarily choose to waive or dissolve their own mental privacy. To illuminate this limitation, the article focuses on a paradigm case made increasingly plausible by contemporary brain–computer interface (BCI) research: the scenario of a group of individuals who voluntarily interconnect their nervous systems into a form of technological "neural telepathy," sharing thoughts, emotions, and mental imagery continuously and without filter. We argue that such a scenario cannot be adequately governed by neurorights alone and demands a complementary framework of neuroduties — legally and morally binding obligations that flow from the embodied, relational nature of the human person and that constrain certain forms of radical voluntary self-transformation. We develop the philosophical foundations of this framework, engaging directly with liberal autonomy theory and with existing legal instruments, and identify three clusters of problems raised by consensual neural interconnection: (1) mental health and the preservation of the bounded self; (2) personal identity, individuation, and the attribution of moral and legal responsibility; and (3) the irreducible social and legal functions of privacy that transcend individual preference.
The article advances the idea of "digital life models" to assist with difficult personal decisions, including those that can change an individual in profound ways. We propose that if personalized AI simulations can model how an individual's life might unfold under different conditions, this could help the individual better understand the likelihoods and expected subjective value of these possible outcomes. It remains to be seen whether these digital life models are technically feasible, given fundamental challenges in modeling psychological complexity, person-environment interactions, and value change. Ethical concerns include data privacy, cultural bias, and the risk of improperly shaping user decisions. Acknowledging these limitations, we argue that digital life models, if successfully developed, could in principle enhance rational and authentic decision-making in the face of potentially transformative change.
This paper examines whether beneficence‑based reasoning can justify the use of psychotropic medications for residents in long‑term aged care facilities who harm or threaten harm to others, a phenomenon described as resident‑to‑resident aggression. I argue that such reasoning works when proportionality‑based steps are integrated, and that this is better achieved through context‑dependent rather than threshold‑bound proportionality. I propose two necessary conditions. The first requires giving proportionate weight to the prospect of benefiting the resident receiving the medication, including both benefits that are intended and benefits that can be reasonably foreseen. The second condition requires reasonable grounds for judging that the intended benefits to the resident are sufficient, when considered on their own, to offset the risks borne by that resident in welfare terms. These two conditions help clarify why the strength of a beneficence-based justification is not determined by intended benefits alone. Its deliberative plausibility depends instead on how intended benefits, reasonably foreseen benefits, and risks relate to one another. When a beneficence‑based justification is not available, I argue that beginning with beneficence‑based considerations still does important moral work. A first‑step beneficence‑based examination clarifies how a more permissive approach, such as institutional harm‑prevention, should be structured. Beneficence‑based reasoning helps identify the constraints needed to prevent institutional harm‑prevention from functioning as a broad restrictive‑licensing tool.
Brain-Computer Interfaces (BCIs) promise to resolve the communication problems of people who are unable to move and speak due to paralysis (Locked-In Syndrome/State; LIS). Current BCI research focuses largely on improving speed and accuracy of BCI-based communication, but this functional perspective fails to capture the existential importance and richness of human communication. In phenomenological studies of human experience, communication is typically theorized as embodied and relational. Here, we propose to use this phenomenological perspective to capture the experience of the communication difficulties for people with LIS and to assess the potential impact of communication solutions such as BCIs. We present the results of a semi-structured phenomenological interview study into experiences of communication of people with LIS and their daily communication partners. We found that people with LIS and daily communication partners are not only at risk of not being able to communicate with each other, but as a result may not see and perceive each other as full persons. Vice versa, access to and richness of bodily and technological communication are enabling factors for the sense of ‘relational personhood’ – the extent to which people acknowledge one another as full persons. These findings pave the way for a new outlook on BCI design and urge to further explore the inclusion of relational personhood in the clinical outcome assessment of BCIs for people with communication impairment and their daily communication partners.
“Study drugs” or “smart drugs,” describe a class of pharmacological agents, usually prescription cognitive stimulants, that are used to increase vigilance (i.e. an individual’s capacity to remain focused on a task in the face of fatigue or distractions) and energy, given their name for their role in educational settings. There is a considerable range of extant positions supporting and critiquing the use of study drugs, with many focused on the role study drugs may play in exacerbating or redressing inequalities. Few accounts, however, do so with attention to whether these interventions work as designed—at least the ones we have today—and how their effectiveness, use, and status within communities might raise ethical issues. We argue, first, that the use of study drugs in educational settings may be unjustified given our available suite of study drugs and current knowledge of their effects. We then argue that the case of study drugs demonstrates outstanding methodological issues in ethical debates about cognitive enhancement, with a focus on how enhancements are purported to overcome existing natural or social inequalities. We conclude with a comment on how we might understand ethically justified cognitive enhancement in the context of our account, as a way to steer debates about the ethics of enhancement that take seriously the empirical challenges ahead.
Psychedelic-assisted therapies are widely described as context-dependent, shaped by “set and setting” and by patients’ broader “matrix” of living conditions. Yet, the field has rarely grappled with the potential ramifications of this context-dependency, including whether an individual’s adverse social circumstances might not only blunt benefits but also amplify harms. This article develops a context-dependent harm hypothesis by analogy to conventional serotonergic antidepressants, for which pre-clinical and clinical evidence supports potential amplification of the harms of detrimental environmental and social contexts. A similar risk is plausible for psychedelics, given that both conventional antidepressants and psychedelic drugs may boost neuroplasticity and environmental sensitivity. This article reviews the limitations of the current psychedelic evidence base that precludes the evaluation of this hypothesis, including the frequent enrollment of socioeconomically advantaged participants and inadequate reporting of social circumstances. It concludes that environment-drug interactions require further attention for scientific, ethical, and clinical purposes. If context-dependent harm exists, adverse living conditions may constitute temporary contraindications until social conditions support positive therapeutic outcomes.
This article adopts a liberal perspective on political legitimacy grounded in the protection of freedom and examines the use of digital technologies as instruments of surveillance, censorship, and manipulation. By taking the cases of Egypt (2011) and Iran (2025) as illustrative entry points, the article highlights repressive strategies in the digital age and shows how political power may operate upon individuals’ psychological processes. The political dynamics associated with cognitive control and repression reveal the limits of the neurorights paradigm. While necessary to safeguard cognitive liberty and mental privacy, neurorights remain defensive individual rights unable to counter the structural asymmetries generated by states and digital platforms. The paper therefore proposes Habeas Mentem, which incorporates and at the same time transcends neurorights by embedding them within a normative, constitutional, and systemic principle, as well as within a broader framework for balancing cognitive power. This framework requires the separation of digital power, antitrust measures, informational pluralism, and critical education, since mental freedom cannot be secured merely by prohibiting intrusions into the mind, but also by constructing the social and institutional conditions that prevent the emergence of mental domination.
Providing ethical care for people living with dementia is essential for safeguarding their rights, dignity, and well-being. However, ethical decision-making in dementia care is complex, often involving tensions between competing moral claims such as autonomy, beneficence, nonmaleficence, justice, and non-discrimination. This scoping review aimed to map and describe the ethical principles, related normative considerations, and ethical frameworks discussed in the scientific literature on dementia care, and to identify how these have been conceptualized. Following the methodological guidance of Arksey and O’Malley (2005), Levac et al. (2010), and Peters et al. (2021, 2022), a comprehensive search across five databases (Web of Science, Scopus, CINAHL, Medline/PubMed, and PsycInfo) identified peer-reviewed papers published in English. In total, 388 papers published between 1981 and 2024 were included. The results revealed a wide range of ethical principles and related normative considerations frequently cited in the literature, although they were typically discussed in an isolated way without reference to other principles or broader ethical frameworks. In contrast, ethical frameworks were mentioned far less often, and when they were, they mostly reflected traditional approaches such as principlism. Four gaps were identified: the plurality of ethical frameworks and limited conceptual integration underpinning dementia care research and practice; insufficient attention to ethnic and socioeconomic diversity among people living with dementia; limited consideration of diverse care settings and geographical regions; and the limited inclusion of people living with dementia themselves in the studies reviewed. These findings underscore the need for conceptually integrative, inclusive, and context-sensitive approaches to dementia care ethics that can better inform both ethical reflection and care practices.
Returning individual research results (IRR) from observational brain studies confronts stakeholders – investigators, participants, families, clinicians, regulators, sponsors – with conflicting ethical, legal, and social considerations. The mismatch between stakeholder standards and expectations leaves investigators uncertain about whether and how to return IRR. We propose to clarify matters by exploring what cognitive liberty, a value salient in neuroscience, portends for the return of IRR from observational neurodegeneration research. IRR specialists have not yet connected cognitive liberty with the return of data from observational studies, despite its similarity to autonomy, a prominent value in this literature. To make this connection, we completed a narrative literature review to inventory the different conceptions of cognitive liberty and assessed their relevance for returning IRR from observational studies of neurodegeneration, with a focus on amyloid-beta imaging results in Alzheimer’s disease. A title/abstract search of PubMed and Web of Science for “cognitive liberty” and a supplementary search returned 16 search results that articulated at least one conception of cognitive liberty, for a total of 32 distinct conceptions. We explored their relevance to returning IRR in two ways. First, we evaluated two of four core conceptions and found that cognitive liberty offers little new explanatory value about why IRR should be returned. Second, we examined two peripheral conceptions – mental integrity and informational self-determination – and judged the latter a defeasible reason to posit investigator duties or participant claims to return amyloid-beta results. Overall, depending on the precise conception, cognitive liberty motivates different stances towards returning amyloid-beta imaging and other IRR.
Advances in Brain-computer interfaces (BCIs) have intensified debates about their alleged capacity to read minds and the implications for mental privacy. Prevailing accounts often assume that brain data can reveal inner mental states, framing privacy as a matter of data protection. Yet experimental demonstrations, such as PIN decoding or semantic reconstruction, show that successful decoding depends on structured tasks, controlled stimuli, and participant cooperation. This paper asks: what does mind reading with BCIs really mean, how should such results be interpreted, and what are the ethical stakes? Drawing on enactivism, we reframe BCI-mediated mind reading not as direct access to hidden thoughts but as an interactional effect co-produced by technology, environment, and user engagement. From this perspective, protecting mental privacy requires more than controlling data: it must safeguard individuals’ autonomy and agency in sense-making and embodied expression. This reframing not only challenges inflated claims about BCI capabilities but also grounds a more robust ethical framework for protecting mental privacy and guiding neurotechnology governance.
Both scientists and philosophers have increasingly focused on the prospect of moral neuroenhancement—the use of neurotechnologies or psychoactive substances to facilitate moral improvement. Recent scholarship distinguishes between two main approaches: direct moral neuroenhancement, which seeks to implant specific moral beliefs, motives, or behaviors, and indirect moral neuroenhancement, which aims to enhance capacities such as moral reasoning, conceptual understanding, or self-control, enabling individuals to arrive at better moral judgments through their own deliberation. Several philosophers have argued that the direct approach is ethically impermissible, as it risks undermining moral autonomy and suppressing valuable disagreement. The indirect approach, by contrast, is taken to avoid these concerns and is widely regarded as ethically permissible or even required. In this paper, we introduce a novel challenge for the project of moral neuroenhancement—the discernibility challenge: for any given intervention, it is exceedingly difficult, if not impossible, to determine whether it truly qualifies as indirect rather than direct. We demonstrate this by examining a range of experimental designs intended to discern between these two types of interventions, arguing that the evidence they produce is ultimately insufficient for this task. The discernibility challenge, we propose, exposes a profound epistemic uncertainty with notable implications for the ethical permissibility of moral neuroenhancement: if we cannot reliably distinguish permissible (indirect) from impermissible (direct) interventions, we lack a crucial epistemic basis for ethically justifying their use.
As neurotechnology and artificial intelligence increasingly converge in NeuroAI systems, concerns about fairness have become central to debates in AI ethics, neuroethics, and law. This article examines a recurring conceptual ambiguity in these debates: the tendency to treat AI algorithmic bias and neurodiscrimination as interchangeable. AI algorithmic bias concerns distortions in data, model design, or deployment that generate systematically unfair outputs. Neurodiscrimination concerns person-centred disadvantage based on neural characteristics or inferences drawn from neurodata. The article further examines the proposed neuroright to protection from algorithmic bias as an important site where this ambiguity becomes especially evident and consequential: although framed as a rights-based response, the proposed right often translates equality concerns into the language of technical system design. The article argues that treating these concepts as interchangeable obscures mechanisms of harm, misallocates responsibility, and misdirects regulatory responses. Through analysis of academic literature, neurorights frameworks, and policy documents, it shows how conceptual slippage could lead to mismatched interventions: technical auditing may be offered where rights-based protection is required, while individual remedies may leave biased systems uncorrected. The article concludes that conceptual clarity is not merely semantic. It is a necessary condition for coherent, accountable, and rights-respecting governance of NeuroAI systems.
Debates on the ethical status of disembodied neural organoids (DNOs) focus on whether, and when, precautionary principles should apply given the uncertain possibility of organoid consciousness. Some advocate applying precautionary principles to DNOs despite uncertainty; others deem such measures premature given current simplicity but foresee their future necessity; while more skeptical views hold that genuinely conscious DNOs remain too remote to warrant present ethical concern. By contrast, Croxford and Bayne [1] defend a more radical skepticism, as they claim that there is little reason to ascribe consciousness not only to current DNOs, but to DNOs as such. They ground this view on a constraint-based approach composed of an Embodiment Constraint (EC) and a Representational Constraint. EC is presented as enjoying broad cross-theoretical support among externalists and internalists, and thus undermining the grounds for a precautionary ethical concern regarding DNOs. We examine EC and argue that, as stated, fails to meet the authors’ own requirement of representing a broad consensus among externalists and internalists.
This study examines cognitive warfare through the lens of critical neuroscience and situates it within peace research as a methodological contribution. Cognitive warfare-understood as the strategic modulation of perception, belief, and decision-making-highlights how the human mind has become a contested domain of contemporary conflict. The weaponization of neuroscience unfolds within sociocultural contexts where empirical findings, theoretical interpretations, and normative commitments are deeply entangled. Drawing on Johan Galtung’s epistemological triad of data, theory, and values, I propose a framework for analyzing cognitive warfare that makes explicit the normative dimensions of neuroscientific knowledge. This perspective highlights not only the weaponization of neuroscience itself, but also of its reputation: exaggerated claims and neuropolitical imaginaries can shape expectations, distort public discourse, and erode trust in science independently of actual technological capabilities. Applying this framework, the study examines resilience and solidarity as two domains in which empirical mechanisms, theoretical interpretations, and normative priors intersect. Cognitive warfare is shown to undermine resilience by destabilizing shared models of reality, while evolutionary and social dynamics demonstrate the long-term advantages of cooperation, reciprocity, and generosity over manipulation and defection. By integrating critical neuroscience with peace research, the article advances an expanded notion of peace as the safeguarding of mental and social autonomy-conditions under which cognition remains free from systematic manipulation and capable of sustaining cooperative forms of life.
Informed consent in psychedelic-assisted services is ethically complex, difficult to implement, and remains largely unstudied and unstandardized. The current study sought expert recommendations on informed consent challenges, best practices and recommendations for supervised psilocybin experiences across various settings. Participants with psilocybin content expertise and psilocybin providers were recruited with purposive sampling. Qualitative interviews on informed consent best practices and recommendations were analyzed using Thematic Analysis. Participants (N = 36; 71
The use of neurocognitive enhancement is controversial. Some social actors and scholars support its application in professions like medicine, while others view it as morally problematic and suggest prohibitive policies. To examine how the public perceives the use of substances for performance enhancement in professional contexts, together with the motivations and consequences of this behavior, we conducted two 2×2×2 between-subjects design scenario-based experiments. The experiments build upon the Agent–Deed–Consequence (ADC) model of moral judgment. A Germany-wide random sample of adults was used in Experiment 1 (N = 1,346) and Experiment 2 (N = 2,161). Both experiments involve scenarios in which a surgeon with either egoistic or altruistic motivation (agent–component) decides either to prepare for the surgery or to use the illegal substance “speed” for performance enhancement (deed–component) before failing or succeeding in surgery (consequence–component). Results show that the illegal substance use condition negatively impacted moral judgment, which statistically accounted for part of the association between the deed and the willingness to undergo surgery. The effect of the Deed on moral judgment and the indirect association of the deed with willingness via moral judgment were moderated by the agent and the consequence components. In summary, this research provides evidence that respondents perceive illegal substances for performance enhancement as morally problematic and that this judgment hinders acceptance of medical services. These results offer support for key assumptions of the ADC model within a healthcare scenario involving illegal enhancement, which could be investigated across a wide range of technologies. They also point to implications for professional bodies to maintain public trust.
Brain-computer interface technology offers potential for non-therapeutic military applications, including enhanced decision-making and human–machine teaming. While ethical implications are frequently debated, empirical research regarding the viewpoints of servicemembers—the end-users facing unique operational risks—is largely missing. This scoping review aims to map the current state of the literature on empirically gathered viewpoints of military personnel toward non-therapeutic BCIs. Three databases (EMBASE, PubMed, and SCOPUS) were searched for articles published in English through October 2025. Supplemental strategies included backward citation tracking and targeted grey literature searches. Eligibility criteria required empirical studies (qualitative, quantitative, or mixed methods) evaluating military servicemember attitudes toward brain-computer interfaces. Two investigators independently performed screening. Only two relevant sources were identified, highlighting a significant data gap. One survey study of 332 international officers found consensus on the need for pre-surgical transparency regarding post-service implant retention. A qualitative study of 13 special forces operators emphasized concerns regarding command pressure, data security, and the absolute necessity of guaranteed long-term, post-service maintenance. There is a near-total absence of empirical data on servicemember perspectives toward military BCIs. The limited evidence reveals foundational concerns regarding trust, accountability, and post-service care. Neglecting end-user viewpoints risks creating ethically misaligned systems. We provide several recommendations for next steps.
