RL-LEGITIMACY Political legitimacy of AI decisions

Making AI Intelligible: Philosophical Foundations · Herman Cappelen; Josh Dever · 2021

“If an AI system tells us that Lucie should not get a mortgage, she is entitled to understand why she should not get a mortgage. To answer the why-question by simply insisting that the decision was made by a reliable but incomprehensible algorithm isn't good enough.”

why coded: Entitlement to intelligible reasons - XAI as metasemantics, pre-figuring the contestability demands · unit #3, pp. 162

Toward a Theory of Justice for Artificial Intelligence (Daedalus 151(2):218-231) · Iason Gabriel · 2022

“These norms entail that the relevant AI systems must meet a certain standard of public justification, support citizens' rights, and promote substantively fair outcomes, something that requires particular attention to the impact they have on the worst-off members of society.”

why coded: Public justification + rights + fair outcomes + worst-off priority as the justice standard for AI · unit #3, pp. 218

Artificial Intelligence, Humanistic Ethics (Daedalus 151(2):232-243) · John Tasioulas · 2022

“How does it feel to contemplate the prospect of a world in which judgments that bear on our deepest interests and moral standing have, at least as their proximate decision-makers, autonomous machines that do not have a share in human solidarity and cannot be held accountable for their decisions in the way that a human judge can?”

why coded: Accountability-capacity as a condition on legitimate decision-makers · unit #5, pp. 237

“[Participation:] These end states could in principle be brought about through a process in which the person who enjoys them is passive: for example, by the government putting a happiness drug into the water supply. Contrary to this passive view, it would stress [active participation in decision-making, individually and as self-governing democratic citizens].”

why coded: Participation as constitutive of well-being and civic dignity - anti-passive-endstate · unit #6, pp. 238

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

“whose voices should be included in the alignment process? And how should we balance input from communities, subject-matter experts and other stakeholders? Individuals do not always hold the most ethical or desirable preferences. Relying exclusively on public inputs might therefore lead to a situation where community rules come into conflict with human rights or other legal considerations.”

why coded: Stakeholder-selection problem again - who is included and on what balance · unit #12, pp. 11

Disagreement, AI alignment, and bargaining · Harry R. Lloyd · 2024

“The project of aligning AIs with human values is arguably more likely to succeed if it can command a broad base of support. But the alignment project is unlikely to command a broad base of support if its intended alignment target only reflects the values of a certain subset of society. [...] It is just bad politics for AI safety proponents to advocate alignment with potentially controversial conceptions of desirable AI behaviour such as total welfare maximisation.”

why coded: Broad-base-of-support argument: alignment needs buy-in to succeed - instrumental legitimacy · unit #3, pp. 1761

Beyond Preferences in AI Alignment · Tan Zhi-Xuan; Micah Carroll; Matija Franklin; Hal… · 2024

“contractualist alignment aims to align AI systems with goals, standards, and principles that are mutually agreed upon by people despite our disparate preferences and values, deriving its normative force from the fair and impartial agreement of relevantly-situated rational actors. [...] AI goals and standards should be justified to each stakeholder, on grounds that none can reasonably reject. Insofar as these AI systems are used to exercise power over others, they should also act in accordance with standards that are not just fair, but legitimate.”

why coded: Legitimacy required where AI exercises power (citing Lazar) · unit #16, pp. 1847

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

“public deliberation aims to generate more granular principles for AI that are tailored to its particular characteristics or deployment scenario, and that are also – to varying degrees – actively affirmed. [...] Efforts to bridge this gap between theory and practice, often reveal a justificatory gap which we believe is better addressed through actual consultation with affected people.”

why coded: Active affirmation as the legitimacy mechanism · unit #15, pp. 1968

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

“When AI operates at a scale that amounts to a form of governance, it becomes subject to norms of governance. In addition to being 'safe' in the narrow sense of 'technically robust,' then, systemically impactful AI must also satisfy standards of public justification and legitimacy (Gabriel & Ghazavi 2022). To the extent that it fails to do so, it poses an authoritarian threat.”

why coded: AI at scale = governance, hence subject to public justification and legitimacy norms · unit #5, pp. 6075

“First, it involves no deliberation between participants [...] These algorithms are too complex for humans to fully comprehend (Burrell 2016), and they do not take their final form until the process is complete. So people cannot deliberate about them until after the fact [...] Second, crowdworkers do not vote, properly speaking. [...] it is not clear, even ex post, how any particular input influences the resultant algorithm.”

why coded: Why crowdsourcing fails politically: no deliberation, no real votes, no input-to-policy transparency · unit #11, pp. 6080

“even if the initial principle selection and specification processes are done through deliberative democratic procedures, the crucial further process of transforming principles into decision-making algorithms is not sufficiently similar to standard democratic procedures to legitimize the system's outputs. And so, even constitutional AI fails on political grounds.”

why coded: Legitimacy breaks at the ML encoding stage even with ideal upstream procedure · unit #15, pp. 6083

“We submit that such decisions are acceptable insofar as (1) they are subject to indirect democratic oversight, and (2) the decision-makers are able to provide reasonable justifications for them, which can in turn enable effective contestation and recourse. [...] Only when combined, then, do these weaker criteria plausibly explain why people have good reason to accept the controversial decisions of unelected arbiters who do not have any special claim to moral expertise.”

why coded: Positive proposal: indirect democratic oversight + justification enabling contestation and recourse · unit #16, pp. 6084

Wide reflective equilibrium in LLM alignment: bridging moral epistemology and AI safety · Matthew Brophy · 2026

“it is not solely the output (moral concordance) that provides moral warrant; MWRE's inherent process virtues – such as systematic error-checking, principled openness to theory change, and comprehensive coherence seeking – supply ethical credentials that are currently underdeveloped in the LLM alignment discourse.”

why coded: Normative warrant from process virtues, not output matching - procedural legitimacy for alignment · unit #11, pp. 9

Agency and alignment: toward a normative architecture for human-AI interaction · Saša Josifović; Jörg Noller · 2026

“This approach leads us to the concept of a normative interface, a design-level structure that facilitates the embedding of AI in action spaces where reasons matter, outputs can be contested, and human agents remain the bearers of final responsibility.”

why coded: Action spaces where reasons matter and outputs can be contested · unit #4, pp. 2

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

“[Lesotho can] contribute to global AI governance through its untapped institutional 'governance resources' of the Pitso (public assembly), Lekhotla (court or council), [the Lekhotla la Baeletsi (council of advisors)], and the Baholisi or Batataisi [guides/mentors].”

why coded: Indigenous deliberative-adjudicative institutions as legitimacy sources · unit #2, pp. 812

Justifications for Democratizing AI Alignment and Their Prospects · Andre Steingrüber; Kevin Baum · 2026

“Two possibilities for what might be inherently bad about epistocratic approaches will be discussed: (i) They give some people illegitimate authority over other people. (ii) Through them some people will be illegitimately coerced by others. We will argue that the latter is the more promising argumentative route.”

why coded: Authority vs coercion as the two illegitimacy risks; coercion the live one for AI · unit #2, pp. 149