TU-LITREV Literature review

Artificial Intelligence, Values, and Alignment · Iason Gabriel · 2020

“The second part of the value alignment question is normative. It asks what values or principles, if any, we ought to encode in artificial agents. Here it is useful to draw a distinction between minimalist and maximalist conceptions of value alignment. The former involves tethering artificial intelligence to some plausible schema of human value and avoiding unsafe outcomes. The latter involves aligning artificial intelligence with the correct or best scheme of human values on a society-wide or global basis.”

why coded: Standard definitional ground for lit review · unit #2, pp. 412

“Jobin et al. (2019) found that 'the underrepresentation of geographic areas such as Africa, South and Central America and Central Asia indicates that global regions are not participating equally in the AI ethics debate, which reveals a power imbalance in international discourse' (396).”

why coded: Global-South underrepresentation in AI ethics discourse · unit #20, pp. 427

STELA: a community-centred approach to norm elicitation for AI alignment · Stevie Bergman; Nahema Marchal; John Mellor; Shak… · 2024

“Members of historically marginalised communities in the US, especially women and people of colour, routinely experience epistemic violence and injustice, where their voices are silenced, their lived experiences and realities are played down or ignored, and they are challenged in their capacity as knowers. [...] they may therefore be especially concerned to avoid replicating this dynamic in their interactions with AI.”

why coded: Epistemic injustice (Fricker lineage) as the explanatory frame for community priorities · unit #8, pp. 10

A matter of principle? AI alignment as the fair treatment of claims · Iason Gabriel; Geoff Keeling · 2025

“A successful answer to this question should ideally meet three criteria. First, it should be relatively complete, providing guidance for AI across multiple domains and scenarios. Second, it should have explanatory value, enhancing our understanding of commonsense ethical judgments about AI and extending their reach. Third, it should possess justificatory power, offering an answer to the alignment question that can be accepted by all who are significantly impacted by AI systems, including people with differing beliefs about value.”

why coded: Three success criteria - usable rubric for evaluating ALL alignment proposals in the lit review · unit #16, pp. 1969

Misalignment or misuse? The AGI alignment tradeoff · Max Hellrigel-Holderbaum; Leonard Dung · 2025

“Other governance proposals, such as mandatory risk assessments before AI deployment by third parties with comprehensive access, clarifying liability for AI harms, compulsory reporting of safety cases, and know-your-customer requirements, may be further uniform improvements [reducing both takeover and misuse risk].”

why coded: Direct bridge from alignment theory to the dissertation's regulation strand · unit #14, pp. 19

AI Alignment: A Comprehensive Survey · Jiaming Ji; Tianyi Qiu; Boyuan Chen; Borong Zhang… · 2025

“we identify four principles as the key objectives of AI alignment: Robustness, Interpretability, Controllability, and Ethicality (RICE). [...] we outline the landscape of current alignment research and decompose them into two key components: forward alignment and backward alignment.”

why coded: RICE + forward/backward cycle - the technical field's self-map · unit #1, pp. 1

Human Value Alignment in AI · Ilias O. Pappas; Polyxeni Vassilakopoulou · 2025

“[Handbook entry: definition and importance of human value alignment in AI; survey of approaches from the human-centered AI / information-systems tradition, incl. value-sensitive design and participatory methods.]”

why coded: HCAI/IS reference-work codification of value alignment - diffusion evidence (tentative) · unit #1, pp. 1

Moral disagreement and the limits of AI value alignment: a dual challenge of epistemic ju… · Nick Schuster; Daniel Kilov · 2025

“For example, Ubuntu philosophy assumes the conceptual and moral priority of the community over the individual, and Confucian philosophy understands rights and obligations primarily in terms of social roles. These views might, therefore, assess projects in value alignment quite differently than we do here. And as AI systems are deployed globally, these and other philosophical perspectives are critical for assessing them appropriately relative to their various contexts of application.”

why coded: Authors' own concession: liberal frame cannot speak for Ubuntu/Confucian perspectives - the invitation for non-Western analysis · unit #3, pp. 6074

Disentangling AI Alignment: A Structured Taxonomy Beyond Safety and Ethics · Kevin Baum · 2026

“An AIA's safety is a strictly monotonically increasing function of its robustness against harmful malfunctions (absent malicious external influences) in foreseeable and intended application contexts.”

why coded: Precise safety definition - malfunction-relative, purpose-agnostic · unit #1, pp. 161

“These AI systems deserve to be called safe—but that does not force us to claim that their use in an unjust war would be morally justified. Stretching the notion of safety to encompass moral permissibility would dilute its conceptual clarity [...] safety does not entail ethicality. [...] while not a conceptual necessity, it is at least a practical regularity that ethicality implies safety.”

why coded: Safety/ethicality entailment analysis - a weapons system can be safe and unethical · unit #3, pp. 164

Agents, Alignment, and the Many Faces of Autonomy · Roberta Fischli; Matija Franklin; Arianna Manzini… · 2026

“Artificial intelligence (AI) systems are becoming more agentic. Capable of predicting, planning and executing actions, AI agents can increasingly perform tasks without human oversight (Knight, 2024; Russell & Norvig, 2016). [...] This puts AI agents at the frontier of AI ethics (Gabriel et al., 2024; Lazar, 2024).”

why coded: Establishes agentic AI as the current frontier of AI ethics - citable framing · unit #1, pp. 2

The Palgrave Handbook on the Ethics of Artificial Intelligence (ed. Gouveia) · Steven S. Gouveia (ed.) · 2026

“[Ch 1: 'Re-aligning Value Alignment: A Metaethical Perspective on AI Ethics' - the handbook opens with the metaethics of value alignment; Ch 18: 'Moral Responsibility Without Moral Agency'; Ch 22: 'Responsible Black Boxes: How Virtue Ethics Can Bridge the Responsibility Gap in AI'; Ch 53: 'Tapping into Basotho Ethical Governance Resources for a Decolonised AI Governance'.]”

why coded: Container: the field's 2026 handbook with responsibility + non-Western clusters mapped · unit #1, pp. 36

Responsibility Attribution for AI-Mediated Damages with Mechanistic Interpretability · Lena Kästner; Johann Cordes; Herbert Zech · 2026

“an unambiguous attribution of responsibility is crucial to ensure legal certainty: for one thing, those affected should be able to assert their rights effectively; for another, various stakeholders should be provided with a clear legal framework to guide their activities. Yet, it remains unclear (a) what conception of causation liability law relies on, (b) how this conception can be utilized to attribute responsibility when human actions rely on the use of opaque AI systems, and (c) how liability for AI-mediated damages should be handled in practice.”

why coded: The three questions (a)-(c) structure the legal side of the responsibility literature · unit #2, pp. 188

Normative Ethics, Artificial Intelligence, and Value Alignment (Dynamic Normativity) · Nicholas Kluge Corrêa · 2026

“[Foreword, M. Gabriel:] his work is not only a contribution to AI ethics but also to a renewed cosmopolitanism of reason: an attempt to articulate a shared normative grammar in a technologically mediated world. [...] it provides empirical clarity about the global moral landscape of AI, showing both where consensus exists and where critical reflection must continue.”

why coded: 'Where consensus exists' - the empirical-consensus framing adjacent to convergentism · unit #2, pp. 1

Understanding the Process of Human-AI Value Alignment · Jack McKinlay; Marina De Vos; Janina A. Hoffmann;… · 2026

“we define value alignment as an ongoing process between humans and autonomous agents that aims to express and implement abstract values in diverse contexts, while managing the cognitive limits of both humans and AI agents and also balancing the conflicting ethical and political demands generated by the values in different groups.”

why coded: The field's consensus definition - cite as the baseline the dissertation refines · unit #1, pp. 1

“Within the reviewed papers, the ethical discussion centred around three Western theories: consequentialism, usually in the form of utilitarianism; deontology; and virtue ethics.”

why coded: Only three Western theories in the entire 172-paper corpus - measured evidence for the Western-canon narrowness · unit #5, pp. 13

“We also observed that non-Western ethical systems and values were neglected in the value alignment research that we analysed. Given that this paper was authored by a Western team, this would have further compounded the effect of Western values on the interpretation of the data. As a result, this paper presents a dominantly Western perspective on the process of value alignment.”

why coded: The review's own confession: non-Western ethics neglected, Western authorship compounds it · unit #10, pp. 28

Tapping into Basotho 'Ethical Governance Resources' for a Decolonised AI Governance (Palg… · Khali Mofuoa · 2026

“governance discourses of these emergent AI technologies have historically been conducted in terms of the perspectives of the Global North. [...] they misleadingly portray the Global North as an epistemological fountain of all knowledge pertaining to the governance discourses and practices of these emergent AI technologies.”

why coded: Global North epistemic monopoly in AI governance named and contested · unit #1, pp. 809

Towards friendly AI: a comprehensive review and new perspectives on human-AI alignment · Qiyang Sun; Yupei Li; Emran Alturki; Sunil Munthu… · 2026

“Friendly Artificial Intelligence (FAI) has been proposed to advocate for more equitable and fair development of AI. [...] This paper addresses these gaps by providing a thorough review of FAI, focusing on theoretical perspectives both for and against its development, and presenting a formal definition [...] Key technical subfields are discussed from the perspectives of eXplainable AI (XAI), privacy, fairness and Affective Computing (AC).”

why coded: FAI umbrella review - engineering-side ethical program mapping · unit #1, pp. 1

The value alignment problem in advisory AI: a systematic literature review · Loukas Triantafyllopoulos; Evgenia Paxinou; Diama… · 2026

“four main approaches can be distinguished. One stream focuses on personalized, preference-based methods [...] A second emphasizes normative or principle-driven strategies that encode ethical, legal, or professional standards directly into advisory processes. A third highlights fairness and cultural adaptation [...] research has concentrated most heavily on preference-based and normative strategies, while fairness- and cognition-oriented perspectives remain less developed.”

why coded: Four-approach map of advisory alignment; fairness/cognition underdeveloped - measured · unit #1, pp. 15