Year and volume: 2026.

Author(s): Avigail Ferdman.

Abstract: Philosophers are grappling with whether artificial intelligence (AI) systems should be permitted to participate in high-stakes moral and political decisions. I draw on Alasdair Macintyre’s political philosophy to resist this possibility. AI cannot qualify as a moral agent or a moral advisor because it cannot participate in reflective deliberations on common goods. Common goods are constitutive of individual goods, since individuals can only reason about their own good as ‘individuals-in-their-social-relationships’, involved in practical activity of common practices. Common practices have goods internal to the practice that can only be achieved in common. This requires cultivating and exercising relational capacities. Increasingly, AI is mediating social relations, doing so in a disembodied manner. AI mediation risks fragmenting social interaction and deskilling the relational capacities necessary for common practices. I demonstrate how this threat might unfold in the context of the common good of knowledge. Knowledge can be perceived as an ‘epistemic commons’: the sharing in the production of knowledge as a common good, with ‘care-taking’ as a fundamental good internal to the practice. This involves the accumulated (often embodied) wisdom of communities of practices sharing a common commitment to listening, thinking, examining, and talking about what is said in the name of knowledge because they care. A tragedy of the epistemic commons occurs when knowledge is pursued for achieving goods external to the practice—money, fame, power, dopamine—things that currently dominate social media practices and AI training. What is more, training future AI on reliable information is insufficient, since AI cannot participate in neither the shared valuing of common knowledge, nor the shared valuing of care-taking practices of common knowledge. Given AI mediation and disembodiment, AI-generated knowledge risks creating distance, distrust and “ir-reciprocity” between humans, undermining the prospects for a ‘common practical life’ necessary both for the common good and humans’ individual good.

Doi: https://doi.org/10.1007/s11245-026-10404-7

Year and volume: 2026 (Vol. 6).

Author(s): David B. Resnik; Mohammad Hosseini; Rico Hauswald.

Abstract: During the initial stage of AI’s incorporation into scientific research, AI systems have functioned predominantly as tools under direct human supervision and control. However, AI incorporation into scientific research is now entering a stage in which AI Agents perform research tasks with partial autonomy while remaining under human supervision and control. In the not-too-distant future, a third stage may arise when autonomous AI systems conduct their own research and generate knowledge without human supervision or control. While the second and third stages of AI-augmented research may offer substantial benefits for science and society, they also create novel ethical issues, including (1) Conducting immoral research that may harm humans and other forms of life; (2) Increasing rate of biased, erroneous and deceptive research; (3) Confidentiality challenges; (4) Overreliance on AI; (5) Diffusion of responsibility and accountability; (6) Deskilling; (7) Job losses; (8) AI-generated research beyond human comprehension; and (9) Erosion of trust. We suggest specific solutions to minimize the negative consequences of these issues and offer proposals for ensuring that incorporation of AI into scientific research supports human values.

Doi: https://doi.org/10.1007/s43681-025-00908-0

Year and volume: 2026 (Vol. 6).

Author(s): Thomas J. Spiegel.

Abstract: Academic philosophers sometimes quip that in the future the only job safe from automation will be that of philosophy professors. However, the current AI revolution has inspired some AI scholars to propose the future establishment of closed-loop AI systems as a type of superintelligent robot that would essentially outperform and replace human scientists Zenil in the future of fundamental science led by generative closed-loop artificial intelligence, arXiv:2307.07522v3, 1-40, 2023; Schmidt in Mach Learn Sci Technol 5(3): 035045, (2024); Kitano in Npj Syst Bio Appl 7: 1–12, (2021). In this paper, I investigate whether, analogously, academic philosophy could be automated by putative, sufficiently advanced future AI, potentially featuring artificial embodiment as robots. To this end, I distinguish two mutually exclusive metaphilosophical conceptions of the nature of philosophy circumscribed as philosophy as a set of propositions (PP) versus philosophy as an activity (PA). Granting AI proponents that – iff artificial general intelligence (AGI), potentially embodied as superintelligent robots is achieved – natural sciences may be fully automated in the future, I argue for the conditional that if PP is true (but not if PA is true), then it is possible that AI can automate philosophy. Additionally, I consider what it would mean to automate philosophy given the current state of LLMs (e.g., the GPT-5 era). Finally, I briefly consider whether it would be preferable for us to have philosophy automated and argue that there are two prima facie reasons why automating philosophy, if possible, might be undesirable: the reason from obsolescence and the reason from ultimate answers.

Doi: https://doi.org/10.1007/s43681-025-00960-w

Year and volume: 2026 (Vol. 207).

Author(s): Giuseppe Pernagallo.

Abstract: Descartes’ cogito ergo sum emerges as an answer to an epistemological crisis: the search for an indubitable foundation in a world vulnerable to error, dreaming, and deception. In the contemporary landscape, however, cognitive activities once experienced as “internal” (reasoning, planning, remembering, writing) are increasingly delegated to artificial intelligence systems. This paper asks whether the Cartesian foundation retains its force under conditions of cognitive delegation, and whether the erosion of first-person thinking threatens the very status of the subject as res cogitans. I argue that the paradox dissolves once we distinguish (i) the epistemic function of the cogito (certainty through self-presence) from (ii) an ontological thesis that would make existence depend on the continuous exercise of thought. Building on a three-level taxonomy of delegation (instrumental support, deliberative suggestion, and agential substitution), a thought experiment involving “total delegation” shows that delegation presupposes minimal acts of intention and assent, while the hypothetical disappearance of conscious thinking undermines certainty rather than existence. The paper then introduces the notion of Cartesian AI to explain why contemporary systems generate the widespread illusion that machines think: by operationalizing a post-Cartesian conception of reason as computation (ratio sine sensu), AI can reproduce the external markers of rational discourse while lacking the first-person self-relation that grounds the cogito. The conclusion is that AI does not refute Descartes but destabilizes the cultural meaning of thinking, shifting attention from metaphysical existence to the normative stakes of agency, authorship, and responsibility.

Doi: https://doi.org/10.1007/s11229-026-05596-9

Year and volume: 2025 (Vol. 92).

Author(s): Adrian K. Yee.

Abstract: Governments and social scientists are increasingly developing machine learning methods to automate the process of identifying terrorists in real time and predict future attacks. However, current operationalizations of “terrorist”’ in artificial intelligence are difficult to justify given three issues that remain neglected: insufficient construct legitimacy, insufficient criterion validity, and insufficient construct validity. I conclude that machine learning methods should be at most used for the identification of singular individuals deemed terrorists and not for identifying possible terrorists from some more general class, nor to predict terrorist attacks more broadly, given intolerably high risks that result from such approaches.

Doi: https://doi.org/10.1017/psa.2024.65

Year and volume: 2019 (Vol. 14).

Author(s): Cameron Buckner.

Abstract: Deep learning is currently the most prominent and widely successful method in artificial intelligence. Despite having played an active role in earlier artificial intelligence and neural network research, philosophers have been largely silent on this technology so far. This is remarkable, given that deep learning neural networks have blown past predicted upper limits on artificial intelligence performance—recognizing complex objects in natural photographs and defeating world champions in strategy games as complex as Go and chess—yet there remains no universally accepted explanation as to why they work so well. This article provides an introduction to these networks as well as an opinionated guidebook on the philosophical significance of their structure and achievements. It argues that deep learning neural networks differ importantly in their structure and mathematical properties from the shallower neural networks that were the subject of so much philosophical reflection in the 1980s and 1990s. The article then explores several different explanations for their success and ends by proposing three areas of inquiry that would benefit from future engagement by philosophers of mind and science.

Doi: https://doi.org/10.1111/phc3.12625

Year and volume: 2024 (Vol. 30).

Author(s): George Kwasi Barimah.

Abstract: In this paper, I develop and defend a moralized conception of epistemic trust in science against a particular kind of non-moral account defended by [Stephen] John (2015, 2018). I suggest that non-epistemic value considerations, non-epistemic norms of communication and affective trust properly characterize the relationship of epistemic trust between scientific experts and non-experts. I argue that it is through a moralized account of epistemic trust in science that we can make sense of the deep-seated moral undertones that are often at play when non-experts (dis)trust science.

Doi: https://doi.org/10.1007/s11948-024-00489-x

Year and volume: 2022 (Vol. 105).

Author(s): 성두현(Doohyun Sung).

Introduction: 최근에 우주바이오 연구개발(R&D)에 관한 논의가 뜨거워짐에 따라 유의미한 투자 및 규제 전략에 대한 요구가 커지고 있다. 본 글은 우주바이오의 미래에 관한 담론에 기여하기 위해 글쓴이의 연구 분야인 과학철학(philosophy of science)의 관점에서 우주바이오 정책의 세부 분야라고 할 수 있는 항공우주의학(aerospace medicine) 규제의 미래를 논하고자 한다.

본 글의 목적은 다음과 같다. 첫째, 본 글은 생명의학 연구자 및 우주바이오 관련 정책 의사결정자(policy decision-maker)를 대상으로 의학 연구(medical research) 및 의약품 규제(pharmaceutical regulation)에 관한 최신 과학철학(philosophy of science) 연구를 소개하고자 한다. 둘째, 본 글은 소개한 문헌을 바탕으로 미래 항공우주의학 규제의 도전 과제를 논의하고자 한다. 마지막으로 본 글은 서술한 도전 과제를 이해하고 논의하는데 있어서 유용한 자원을 소개하고자 한다.

본 글의 순서는 다음과 같다. 우선 본 글은 먼저 항공우주의학이라는 분야의 특성을 간략하게 논하고자 한다. 과학기술 분야의 모든 특성을 빠짐없이 논하는 것은 매우 난해하므로 본 글은 의학 연구 및 제약규제의 관점에서 중요하다고 판단되는 특성(연구방법론 및 치료방식) 위주로 논한다. 다음으로 본 글은 의료연구 및 제약규제에 관한 과학철학 문헌을 바탕으로 미래 항공우주의학 규제가 극복해야 하는 도전 과제를 소개한다. 마지막으로 본 글은 소개한 도전 과제에 관한 정책담론에 보탬이 될 수 있는 자원을 소개하고자 한다.

본 글은 논의를 시작하기에 앞서서 다음을 강조하고자 한다. 첫째, 본 글의 목적은 특정 정책안 혹은 제도 설계를 제시하는 것이 아니라 항공우주의학 규제를 위한 정책안을 제시하는데 있어서 고찰해야 하는 전략적 난제를 소개하는 것이다. 둘째, 본 글은 미국 식품의약국(FDA)으로 대표되는 미국 제약규제 맥락을 대상으로 한 과학철학 문헌을 논의의 바탕으로 삼는다. 이는 기존 과학철학 연구가 미국 제약규제를 주요 논의의 대상으로 삼아왔기 때문이다. 다만 본 글은 소개하는 전략적 도전 과제가 한국의 제약규제의 맥락에서도 고찰되어야 한다고 주장한다.

URL: https://www.bioin.or.kr/board.do?num=318730&cmd=view&bid=report

Year and volume: 2023 (Vol. 90).

Author(s): Eamon Duede.

Abstract: While philosophers have focused on epistemological and ethical challenges of using artificial intelligence (AI) in science, scientists have focused largely on opportunities. I argue that this disconnect between philosophical pessimism and scientific optimism is driven by failures to critically examine the practice of AI-infused science. To appreciate the epistemic justification for AI-powered breakthroughs, philosophers must analyze the role of AI as part of a wider process of discovery. I demonstrate the importance of this with two cases from the scientific literature, and show that epistemic opacity need not diminish AI’s capacity to lead scientists to significant and justifiable breakthroughs.

Doi: https://doi.org/10.1017/psa.2023.8

Year and volume: 2026 (Vol. 45).

Author(s): Nina Lauharatanahirun; Jason A. Aimone; Jeffrey Braxton Gately.

Abstract: Research has long documented how decision-making in risky environments differs between environments where the probabilities of uncertain outcomes are known and where the probabilities are unknown, the latter often referred to as “ambiguous” environments. Yet, there is a dearth of research examining how decisions may be affected by the source responsible for the distribution of uncertain outcomes. The source responsible for generating distributions of uncertain outcomes may be generated by another person (i.e., is social in nature) or by a nonsocial probabilistic mechanism. While a few studies examine how the source responsible for uncertain outcomes affects decisions when probabilities are known, the present study extends prior research to the realm of ambiguity by testing how the source of uncertainty affects both decisions when probabilities are fully known and when probabilities are partially unknown using a within-subjects experimental design. We calculate a general measurement of Social Risk Sensitivity to capture how individuals differ in their sensitivity across three uncertainty environments: risk with no ambiguity, risk with low ambiguity, and risk with high ambiguity. We find evidence showing strong correlations between Social Risk Sensitivity across all three levels of ambiguity. Our results corroborate the previous literature regarding ambiguity effects on decision-making behavior and extend prior work for the first time in this area by demonstrating that individual decisions are shaped by their individual sensitivity to the source from which uncertainty is derived.

Doi: https://doi.org/10.1007/s11229-025-05044-0