Although large language models (LLMs) are increasingly implicated in interpersonal and societal decision-making, their ability to navigate explicit conflicts between legitimately different cultural value systems remains largely unexamined. Existing benchmarks predominantly target cultural knowledge (CulturalBench), value prediction (WorldValuesBench), or single-axis bias diagnostics (CDEval); none evaluate how LLMs adjudicate when multiple culturally grounded values directly clash. We address this gap with CCD-Bench, a benchmark that assesses LLM decision-making under cross-cultural value conflict. CCD-Bench comprises 2,182 open-ended dilemmas spanning seven domains, each paired with ten anonymized response options corresponding to the ten GLOBE cultural clusters. These dilemmas are presented using a stratified Latin square to mitigate ordering effects. We evaluate 17 non-reasoning LLMs. Models disproportionately prefer Nordic Europe (mean 20.2 percent) and Germanic Europe (12.4 percent), while options for Eastern Europe and the Middle East and North Africa are underrepresented (5.6 to 5.8 percent). Although 87.9 percent of rationales reference multiple GLOBE dimensions, this pluralism is superficial: models recombine Future Orientation and Performance Orientation, and rarely ground choices in Assertiveness or Gender Egalitarianism (both under 3 percent). Ordering effects are negligible (Cramer's V less than 0.10), and symmetrized KL divergence shows clustering by developer lineage rather than geography. These patterns suggest that current alignment pipelines promote a consensus-oriented worldview that underserves scenarios demanding power negotiation, rights-based reasoning, or gender-aware analysis. CCD-Bench shifts evaluation beyond isolated bias detection toward pluralistic decision making and highlights the need for alignment strategies that substantively engage diverse worldviews.

Large Language Models (LLMs) alignment methods have been credited with the commercial success of products like ChatGPT, given their role in steering LLMs towards user-friendly outputs. However, current alignment techniques predominantly mirror the normative preferences of a narrow reference group, effectively imposing their values on a wide user base. Drawing on theories of the power/knowledge nexus, this work argues that current alignment practices centralise control over knowledge production and governance within already influential institutions. To counter this, we propose decentralising alignment through three characteristics: context, pluralism, and participation. Furthermore, this paper demonstrates the critical importance of delineating the context-of-use when shaping alignment practices by grounding each of these features in concrete use cases. This work makes the following contributions: (1) highlighting the role of context, pluralism, and participation in decentralising alignment; (2) providing concrete examples to illustrate these strategies; and (3) demonstrating the nuanced requirements associated with applying alignment across different contexts of use.

The conceptual framework proposed in this paper centers on the development of a deliberative moral reasoning system - one designed to process complex moral situations by generating, filtering, and weighing normative arguments drawn from diverse ethical perspectives. While the framework is rooted in Machine Ethics, it also makes a substantive contribution to Value Alignment by outlining a system architecture that links structured moral reasoning to action under time constraints. Grounded in normative moral pluralism, this system is not constructed to imitate behavior but is built on reason-sensitive deliberation over structured moral content in a transparent and principled manner. Beyond its role as a deliberative system, it also serves as the conceptual foundation for a novel two-level architecture: functioning as a moral reasoning teacher envisioned to train faster models that support real-time responsiveness without reproducing the full structure of deliberative reasoning. Together, the deliberative and intuitive components are designed to enable both deep reflection and responsive action. A key design feature is the dual-hybrid structure: a universal layer that defines a moral threshold through top-down and bottom-up learning, and a local layer that learns to weigh competing considerations in context while integrating culturally specific normative content, so long as it remains within the universal threshold. By extending the notion of moral complexity to include not only conflicting beliefs but also multifactorial dilemmas, multiple stakeholders, and the integration of non-moral considerations, the framework aims to support morally grounded decision-making in realistic, high-stakes contexts.

Large language models (LLMs) generate diverse, situated, persuasive texts from a plurality of potential perspectives, influenced heavily by their prompts and training data. As part of LLM adoption, we seek to characterize - and ideally, manage - the socio-cultural values that they express, for reasons of safety, accuracy, inclusion, and cultural fidelity. We present a validated approach to automatically (1) extracting heterogeneous latent value propositions from texts, (2) assessing resonance and conflict of values with texts, and (3) combining these operations to characterize the pluralistic value alignment of human-sourced and LLM-sourced textual data.

While existing alignment paradigms have been integral in developing large language models (LLMs), LLMs often learn an averaged human preference and struggle to model diverse preferences across cultures, demographics, and communities. We propose Modular Pluralism, a modular framework based on multi-LLM collaboration for pluralistic alignment: it "plugs into" a base LLM a pool of smaller but specialized community LMs, where models collaborate in distinct modes to flexibility support three modes of pluralism: Overton, steerable, and distributional. Modular Pluralism is uniquely compatible with black-box LLMs and offers the modular control of adding new community LMs for previously underrepresented communities. We evaluate Modular Pluralism with six tasks and four datasets featuring questions/instructions with value-laden and perspective-informed responses. Extensive experiments demonstrate that Modular Pluralism advances the three pluralism objectives across six black-box and open-source LLMs. Further analysis reveals that LLMs are generally faithful to the inputs from smaller community LLMs, allowing seamless patching by adding a new community LM to better cover previously underrepresented communities.

With increased power and prevalence of AI systems, it is ever more critical that AI systems are designed to serve all, i.e., people with diverse values and perspectives. However, aligning models to serve pluralistic human values remains an open research question. In this piece, we propose a roadmap to pluralistic alignment, specifically using language models as a test bed. We identify and formalize three possible ways to define and operationalize pluralism in AI systems: 1) Overton pluralistic models that present a spectrum of reasonable responses; 2) Steerably pluralistic models that can steer to reflect certain perspectives; and 3) Distributionally pluralistic models that are well-calibrated to a given population in distribution. We also propose and formalize three possible classes of pluralistic benchmarks: 1) Multi-objective benchmarks, 2) Trade-off steerable benchmarks, which incentivize models to steer to arbitrary trade-offs, and 3) Jury-pluralistic benchmarks which explicitly model diverse human ratings. We use this framework to argue that current alignment techniques may be fundamentally limited for pluralistic AI; indeed, we highlight empirical evidence, both from our own experiments and from other work, that standard alignment procedures might reduce distributional pluralism in models, motivating the need for further research on pluralistic alignment.

Work in AI ethics and fairness has made much progress in regulating LLMs to reflect certain values, such as fairness, truth, and diversity. However, it has taken the problem of how LLMs might 'mean' anything at all for granted. Without addressing this, it is not clear what imbuing LLMs with such values even means. In response, we provide a general theory of meaning that extends beyond humans. We use this theory to explicate the precise nature of LLMs as meaning-agents. We suggest that the LLM, by virtue of its position as a meaning-agent, already grasps the constructions of human society (e.g. morality, gender, and race) in concept. Consequently, under certain ethical frameworks, currently popular methods for model alignment are limited at best and counterproductive at worst. Moreover, unaligned models may help us better develop our moral and social philosophy.

Texas residents share how familiar they are with artificial intelligence on a scale from one to 10 and detailed how much they use it each day. The "Godfather of A.I.," Geoffrey Hinton, quit Google out of fear that his former employer intends to deploy artificial intelligence in ways that will harm human beings. "It is hard to see how you can prevent the bad actors from using it for bad things," Hinton recently told The New York Times. But stomping out the door does nothing to atone for his own actions, and it certainly does nothing to protect conservatives – who are the primary target of A.I. programmers – from being canceled. Here are five things to know as the battle over A.I. turns hot: Elon Musk recently revealed that Google co-founder Larry Page and other Silicon Valley leaders want AI to establish a "digital god" that "would understand everything in the world.

Artificial Intelligence (AI) started out with an ambition to reproduce the human mind, but, as the sheer scale of that ambition became manifest, it quickly retreated into either studying specialized intelligent behaviours, or proposing over-arching architectural concepts for interfacing specialized intelligent behaviour components, conceived of as agents in a kind of organization. This agent-based modeling paradigm, in turn, proves to have interesting applications in understanding, simulating, and predicting the behaviour of social and legal structures on an aggregate level. For these reasons, this chapter examines a number of relevant cross-cutting concerns, conceptualizations, modeling problems and design challenges in large-scale distributed Artificial Intelligence, as well as in institutional systems, and identifies potential grounds for novel advances.

As a driver, have you ever asked yourself whether to make left turns? Unprotected left turns, that is, left turns with oncoming traffic, are among the most difficult and dangerous driving maneuvers. Although the risk of each individual left turn is negligible, if you are designing the behavior of a large fleet of self-driving cars, small individual risks add up to a significant number of expected injuries in the aggregate. Whether a fleet of cars should make left turns is a question that any developer of self-driving cars and any designer of mapping and routing applications faces today. A more general issue is at stake here: the decision of whether to make left turns involves a trade-off between safety and mobility (the time it takes to get to a destination).