Citation: Durán, J.M. (Forthcoming). Against epistemic transparency of algorithms. Philosophy of Science. https://doi.org/10.1017/psa.2026.10246.

Abstract: Epistemic transparency is often proposed as a solution to algorithmic opacity, wherein revealing the inner logic of an algorithm provides reasons or supporting evidence for the justification of its outputs. I argue that transparency is defective in belief formation and thus inadequate as an epistemology of algorithms. Two objections are developed: transparency regress and bootstrapping, both rooted in transparency’s status as a time-sliced, “outsourced” inferentialist epistemology. While weaker forms—such as contextual transparency—are briefly considered, the central argument remains that transparency fails to provide justification and should be abandoned as an epistemology of algorithms.

Citation: Lu, W., & S. Sun. (2026). The failure of the self-control paradigm: understanding autonomy risks in Human-AI Interaction. Synthese, 208. https://doi.org/10.1007/s11229-026-05713-8.

Abstract: As the integration of Artificial Intelligence (AI) into learning, work, and daily life intensifies, the practice of outsourcing cognitive tasks to AI systems has raised concerns regarding the erosion of human control and autonomy. This article employs an immanent critique to challenge the view that attributes autonomy risks solely to implementation failures of control in practice. Instead, it reframes the theoretical roots of these risks as a systemic failure of the self-control paradigm itself. Practically, AI systems distort user control over cognition and decision-making and contribute to deskilling—autonomy challenges that cannot be resolved merely by strengthening self-control. Theoretically, the self-control paradigm fails to account for adaptive preference formation due to its overemphasis on users’ deliberative capacities and neglects of the constitutive role of algorithmic environments. To preserve and support human autonomy, we argue for the necessity of transcending the narrow conception of self-control in favor of an alternative paradigm of autonomy within Human-AI Interaction.

Citation: Achella, S., & J.G. Ganascia. (2026). Introduction: Rethinking Human-Centredness: Toward an Ecological and Technological Ethics of Interdependence. Topoi (2026). https://doi.org/10.1007/s11245-026-10469-4.

Citation: Ludwig, K. (2026). The ethics of corporate belief. Synthese, 208. https://doi.org/10.1007/s11229-026-05679-7.

Abstract: I argue for a deflationary account of corporate belief on which their attribution is grounded in the official and intentional actions of the corporation’s operative members. I outline an ethics of individual belief focusing on responsibility for belief formation in contexts in which beliefs guide actions that may bring about harm. In light of the deflationary account of corporate attitudes, I consider how to project the account to the corporate case. I argue that the division of role responsibilities, especially proxy roles, in organizations induces a division of moral labor. On this basis, I introduce a framework for identifying roles that, relative to particular actions, are responsibility magnets and responsibility shelters for institutional role players. I identify epistemic and oversight roles as responsibility magnets for corporate belief.

Year and volume: 2026 (Vol. 207).

Author(s): Gregory Wheeler.

Abstract: Bounded rationality is typically understood as a concession to human cognitive limitations, a departure from an ideal coherent in principle if unattainable in practice. I argue this gets the relationship backwards. Unbounded rationality is a physical impossibility, and its attendant normative standards—Bayesian updating, sure loss avoidance, expected utility maximization, logical closure—are techniques for favorable circumstances when resources permit, not ideals from which mortals regrettably fall short. The argument rests on the physics of computation: any information-processing system incurs irreducible costs in energy and time. Three independent lines of support establish this conclusion. The first runs through Landauer’s principle. The second draws on Wolpert’s stochastic thermodynamics framework, extended by Kolchinsky and Wolpert to Turing machines, where thermodynamic costs track algorithmic complexity. The third draws on quantum-mechanical and relativistic bounds that fix finite ceilings on the operations any physical system can perform in a given region of space and time. These constraints bind all physical systems, natural or artificial. Coherence conditions like Savage’s axioms and Bayesian probability evaluate global states, not local procedures. No finite physical process can construct, verify, or maintain global coherence over a realistic state space. They can still function diagnostically, as devices for flagging departures worth explaining; what they cannot do is serve as action-guiding norms from which bounded reasoners fall short. Normative theories that presuppose unbounded rationality demand the physically impossible. Bounded rationality is not a departure from ideal rationality. It is the only kind there is.

Doi: https://doi.org/10.1007/s11229-026-05625-7

Year and volume: 2021 (Vol. 35).

Author(s): Petar Bodlović.

Abstract: The study of presumptions has intensified in argumentation theory over the last years. Although scholars put forward different accounts, they mostly agree that presumptions can be studied in deliberative and epistemic contexts, have distinct contextual functions (guiding decisions vs. acquiring information), and promote different kinds of goals (non-epistemic vs. epistemic). Accordingly, there are “practical” and “cognitive” presumptions. In this paper, I show that the differences between practical and cognitive presumptions go far beyond contextual considerations. The central aim is to explore Nicholas Rescher’s contention that both types of presumptions have a closely analogous pragmatic function, i.e., that practical and cognitive presumptions are made to avoid greater harm in circumstances of epistemic uncertainty. By comparing schemes of practical and cognitive reasoning, I show that Rescher’s contention requires qualifications. Moreover, not only do practical and cognitive presumptions have distinct pragmatic functions, but they also perform different dialogical functions (enabling progress vs. preventing regress) and, in some circumstances, cannot be defeated by the same kinds of evidence. Hence, I conclude that the two classes of presumptions merit distinct treatment in argumentation theory.

Doi: https://doi.org/10.1007/s10503-020-09536-w

Year and volume: 2026 (Vol. 207).

Author(s): Tanja Rechnitzer.

Abstract: Reflective Equilibrium (RE) is often cited as an influential philosophical method but faces a range of objections, including that it is overly demanding, vacuous, or methodologically irrelevant. This paper identifies a fundamental issue that complicates both the evaluation of RE and broader discussions of philosophical methodology: the lack of clarity about what constitutes a method in philosophy, what we expect from such a method, and what kinds of objections against it are legitimate. By distinguishing between epistemology, methodology, and method, I identify three levels at which RE might be located. The term ‘reflective equilibrium’ is often used to encompass all three, obscuring important differences. I argue that while RE can be conceptualized at each level, for guiding philosophical inquiry, it is most plausible and fruitful to understand it primarily as a methodology rather than a method. Beyond debates about RE, the proposed tripartite distinction may also serve as a helpful framework for reflecting more systematically on how epistemological assumptions, methodological commitments, and the use of methods interact in philosophical inquiry.

Doi: https://doi.org/10.1007/s11229-026-05612-y

Year and volume: 2026 (Vol. 6).

Author(s): Raphael Max.

Abstract: AI ethics has become a global industry of principles. Across governments, intergovernmental bodies, research institutes, and firms, guideline documents repeatedly endorse fairness, transparency, accountability, privacy, and human dignity. Yet convergence on labels is often mistaken for convergence on normative content. This paper argues that much of the current discourse remains normatively influential while being insufficiently action-guiding. It develops two claims. First, normative concepts are genuinely useful only when they are determinate enough to support critique: they must specify what would count as satisfying them and what would count as failing them, and they must confront trade-offs rather than implying a moral free lunch in which all values can be maximized simultaneously. Second, even conceptually clear normativity is incomplete if it treats implementation as an afterthought. Normative claims about sociotechnical systems implicitly rely on assumptions about feasibility, institutional incentives, measurement, and human response. When these background assumptions remain unspecified or idealized, ethical guidance becomes aspirational and can even be normatively irresponsible. The paper therefore articulates minimal conditions for AI-ethical discourse that is criticizable and responsibly connected to implementation, without reducing ethics to technocracy or compliance checklists.

Doi: https://doi.org/10.1007/s43681-026-01188-y

Year and volume: 2026.

Author(s): Mandi Astola; Eugen Popa.

Abstract: The concept of “bullshit” (Frankfurt H (1986) On Bullshit. Raritan.) is a well-established philosophical tool for criticizing speech that lacks a serious relationship to truth conditions or evidence. While the concept has been employed primarily to evaluate discourse, we argue in this paper that bullshit is also relevant to the assessment of technology. We introduce the concept of technological bullshit as a means of critically reflecting on artifacts and their societal construction. As a case in point for the analysis of technological bullshit, we focus on a two-wheeler that has caused some recent controversy in the field of urban mobility: the fatbike. We illustrate the identification and analysis of technological bullshit by offering a case study on the pragmatics of fatbikes, i.e., how they are made, used, and talked about. Based on this, and some additional examples, we finally characterize different forms of technological bullshit and conclude with a note on how technological bullshit can be used as a form of critique.

Doi: https://doi.org/10.1007/s11245-026-10425-2

Year and volume: 2026 (Vol. 207).

Author(s): Gregory Wheeler.

Abstract: Bounded rationality is typically understood as a concession to human cognitive limitations, a departure from an ideal coherent in principle if unattainable in practice. I argue this gets the relationship backwards. Unbounded rationality is a physical impossibility, and its attendant normative standards—Bayesian updating, sure loss avoidance, expected utility maximization, logical closure—are techniques for favorable circumstances when resources permit, not ideals from which mortals regrettably fall short. The argument rests on the physics of computation: any information-processing system incurs irreducible costs in energy and time. Three independent lines of support establish this conclusion. The first runs through Landauer’s principle. The second draws on Wolpert’s stochastic thermodynamics framework, extended by Kolchinsky and Wolpert to Turing machines, where thermodynamic costs track algorithmic complexity. The third draws on quantum-mechanical and relativistic bounds that fix finite ceilings on the operations any physical system can perform in a given region of space and time. These constraints bind all physical systems, natural or artificial. Coherence conditions like Savage’s axioms and Bayesian probability evaluate global states, not local procedures. No finite physical process can construct, verify, or maintain global coherence over a realistic state space. They can still function diagnostically, as devices for flagging departures worth explaining; what they cannot do is serve as action-guiding norms from which bounded reasoners fall short. Normative theories that presuppose unbounded rationality demand the physically impossible. Bounded rationality is not a departure from ideal rationality. It is the only kind there is.

Doi: https://doi.org/10.1007/s11229-026-05625-7