| Item Type | Journal Article |
|---|---|
| Author | Daron Acemoglu |
| Author | Nicolás Ajzenman |
| Author | Cevat Giray Aksoy |
| Author | Martin Fiszbein |
| Author | Carlos Molina |
| Abstract | Using large-scale survey data covering more than 110 countries and exploiting within-country variation across cohorts and surveys, we show that individuals with longer exposure to democracy display stronger support for democratic institutions, and that this effect is almost entirely driven by exposure to democracies with successful performance in terms of economic growth, control of corruption, peace and political stability, and public goods provision. Across a variety of specifications, estimation methods, and samples, the results are robust, and the timing and nature of the effects are consistent with our interpretation. We also present suggestive evidence that democratic institutions that receive support from their citizens perform better in the face of negative shocks. |
| Date | 2025-03-06 |
| Language | en |
| Library Catalog | DOI.org (Crossref) |
| URL | https://academic.oup.com/restud/article/92/2/621/7675443 |
| Accessed | 4/11/2025, 11:47:00 AM |
| Rights | https://academic.oup.com/pages/standard-publication-reuse-rights |
| Volume | 92 |
| Pages | 621-655 |
| Publication | Review of Economic Studies |
| DOI | 10.1093/restud/rdae051 |
| Issue | 2 |
| ISSN | 0034-6527, 1467-937X |
| Date Added | 4/14/2025, 9:44:59 PM |
| Modified | 4/14/2025, 9:44:59 PM |
| Item Type | Journal Article |
|---|---|
| Author | Xabier E. Barandiaran |
| Author | Marta Pérez-Verdugo |
| Abstract | This paper introduces the concept of “generative midtended cognition”, that explores the integration of generative AI technologies with human cognitive processes. The term “generative” reflects AI’s ability to iteratively produce structured outputs, while “midtended” captures the potential hybrid (human-AI) nature of the process. It stands between traditional conceptions of intended creation, understood as steered or directed from within, and extended processes that bring exo-biological processes into the creative process. We examine the working of current generative technologies (based on multimodal transformer architectures typical of large language models like ChatGPT) to explain how they can transform human cognitive agency beyond what the conceptual resources of standard theories of extended cognition can capture. We suggest that the type of cognitive activity typical of the coupling between a human and generative technologies is closer (but not equivalent) to social cognition than to classical extended cognitive paradigms. Yet, it deserves a specific treatment. We provide an explicit definition of generative midtended cognition in which we treat interventions by AI systems as constitutive of the agent’s intentional creative processes. Furthermore, we distinguish two dimensions of generative hybrid creativity: 1. Width: captures the sensitivity of the context of the generative process (from the single letter to the whole historical and surrounding data), 2. Depth: captures the granularity of iteration loops involved in the process. Generative midtended cognition stands in the middle depth between conversational forms of cognition in which complete utterances or creative units are exchanged, and micro-cognitive (e.g. neural) subpersonal processes. Finally, the paper discusses the potential risks and benefits of widespread generative AI adoption, including the challenges of authenticity, generative power asymmetry, and creative boost or atrophy. |
| Date | 2025-03-27 |
| Language | en |
| Short Title | Generative midtended cognition and Artificial Intelligence |
| Library Catalog | Springer Link |
| URL | https://doi.org/10.1007/s11229-025-04961-4 |
| Accessed | 4/13/2025, 6:30:42 PM |
| Volume | 205 |
| Pages | 137 |
| Publication | Synthese |
| DOI | 10.1007/s11229-025-04961-4 |
| Issue | 4 |
| Journal Abbr | Synthese |
| ISSN | 1573-0964 |
| Date Added | 4/14/2025, 9:45:03 PM |
| Modified | 4/14/2025, 9:45:03 PM |
| Item Type | Journal Article |
|---|---|
| Author | Herman Cappelen |
| Author | Josh Dever |
| Author | John Hawthorne |
| Abstract | This paper presents an argument that certain AI safety measures, rather thanmitigating existential risk, may instead exacerbate it. Under certain key assumptions -the inevitability of AI failure, the expected correlation between an AI system's power atthe point of failure and the severity of the resulting harm, and the tendency of safetymeasures to enable AI systems to become more powerful before failing - safety effortshave negative expected utility. The paper examines three response strategies:Optimism, Mitigation, and Holism. Each faces challenges stemming from intrinsicfeatures of the AI safety landscape that we term Bottlenecking, the Perfection Barrier,and Equilibrium Fluctuation. The surprising robustness of the argument forces a reexaminationof core assumptions around AI safety and points to several avenues forfurther research. |
| Date | 2025-03-06 |
| Language | en |
| Short Title | AI safety |
| Library Catalog | Springer Link |
| URL | https://doi.org/10.1007/s11098-025-02297-w |
| Accessed | 4/14/2025, 11:11:49 AM |
| Publication | Philosophical Studies |
| DOI | 10.1007/s11098-025-02297-w |
| Journal Abbr | Philos Stud |
| ISSN | 1573-0883 |
| Date Added | 4/14/2025, 9:45:03 PM |
| Modified | 4/14/2025, 9:45:03 PM |
| Item Type | Journal Article |
|---|---|
| Author | Mark Coeckelbergh |
| Abstract | In the ethics of artificial intelligence literature, there is increasing attention to knowledge-related issues such as explainability, bias, and epistemic bubbles. This paper investigates epistemic problems raised by AI and their normative implications through the lens of the concept of epistemic agency. How is epistemic agency impacted by AI? The paper argues that the use of artificial intelligence and data science, while offering more information, risks to influence the formation and revision of our beliefs in ways that diminish our epistemic agency. Using examples of someone who struggles to revise her beliefs, the paper discusses several intended and non-intended influences. It analyses these problems by engaging with the literature on epistemic agency and on the political epistemology of digital technologies, discussing the ethical and political consequences, and indicates some directions for technology and education policy. |
| Short Title | AI and Epistemic Agency |
| Library Catalog | Taylor and Francis+NEJM |
| URL | https://doi.org/10.1080/02691728.2025.2466164 |
| Accessed | 4/13/2025, 6:24:16 PM |
| Extra | Publisher: Routledge _eprint: https://doi.org/10.1080/02691728.2025.2466164 |
| Volume | 0 |
| Pages | 1-13 |
| Publication | Social Epistemology |
| DOI | 10.1080/02691728.2025.2466164 |
| Issue | 0 |
| ISSN | 0269-1728 |
| Date Added | 4/14/2025, 9:45:03 PM |
| Modified | 4/14/2025, 9:45:03 PM |
| Item Type | Preprint |
|---|---|
| Author | Jillian Fisher |
| Author | Ruth E. Appel |
| Author | Chan Young Park |
| Author | Yujin Potter |
| Author | Liwei Jiang |
| Author | Taylor Sorensen |
| Author | Shangbin Feng |
| Author | Yulia Tsvetkov |
| Author | Margaret E. Roberts |
| Author | Jennifer Pan |
| Author | Dawn Song |
| Author | Yejin Choi |
| Abstract | AI systems often exhibit political bias, influencing users' opinions and decision-making. While political neutrality-defined as the absence of bias-is often seen as an ideal solution for fairness and safety, this position paper argues that true political neutrality is neither feasible nor universally desirable due to its subjective nature and the biases inherent in AI training data, algorithms, and user interactions. However, inspired by Joseph Raz's philosophical insight that "neutrality [...] can be a matter of degree" (Raz, 1986), we argue that striving for some neutrality remains essential for promoting balanced AI interactions and mitigating user manipulation. Therefore, we use the term "approximation" of political neutrality to shift the focus from unattainable absolutes to achievable, practical proxies. We propose eight techniques for approximating neutrality across three levels of conceptualizing AI, examining their trade-offs and implementation strategies. In addition, we explore two concrete applications of these approximations to illustrate their practicality. Finally, we assess our framework on current large language models (LLMs) at the output level, providing a demonstration of how it can be evaluated. This work seeks to advance nuanced discussions of political neutrality in AI and promote the development of responsible, aligned language models. |
| Date | 2025-02-18 |
| Library Catalog | arXiv.org |
| URL | http://arxiv.org/abs/2503.05728 |
| Accessed | 4/11/2025, 10:31:20 AM |
| Extra | arXiv:2503.05728 [cs] |
| DOI | 10.48550/arXiv.2503.05728 |
| Repository | arXiv |
| Archive ID | arXiv:2503.05728 |
| Date Added | 4/14/2025, 9:44:59 PM |
| Modified | 4/14/2025, 9:44:59 PM |
Comment: Code: https://github.com/jfisher52/Approximation_Political_Neutrality
| Item Type | Journal Article |
|---|---|
| Author | Iason Gabriel |
| Author | Geoff Keeling |
| Abstract | The normative challenge of AI alignment centres upon what goals or values ought to be encoded in AI systems to govern their behaviour. A number of answers have been proposed, including the notion that AI must be aligned with human intentions or that it should aim to be helpful, honest and harmless. Nonetheless, both accounts suffer from critical weaknesses. On the one hand, they are incomplete: neither specification provides adequate guidance to AI systems, deployed across various domains with multiple parties. On the other hand, the justification for these approaches is questionable and, we argue, of the wrong kind. More specifically, neither approach takes seriously the need to justify the operation of AI systems to those affected by their actions – or what this means for pluralistic societies where people have different underlying beliefs about value. To address these limitations, we propose an alternative account of AI alignment that focuses on fair processes. We argue that principles that are the product of these processes are the appropriate target for alignment. This approach can meet the necessary standard of public justification, generate a fuller set of principles for AI that are sensitive to variation in context, and has explanatory power insofar as it makes sense of our intuitions about AI systems and points to a number of hitherto underappreciated ways in which an AI system may cease to be aligned. |
| Date | 2025-03-30 |
| Language | en |
| Short Title | A matter of principle? |
| Library Catalog | DOI.org (Crossref) |
| URL | https://link.springer.com/10.1007/s11098-025-02300-4 |
| Accessed | 4/2/2025, 12:00:37 PM |
| Publication | Philosophical Studies |
| DOI | 10.1007/s11098-025-02300-4 |
| Journal Abbr | Philos Stud |
| ISSN | 0031-8116, 1573-0883 |
| Date Added | 4/2/2025, 2:13:04 PM |
| Modified | 4/2/2025, 2:13:04 PM |
| Item Type | Journal Article |
|---|---|
| Author | Atoosa Kasirzadeh |
| Abstract | The conventional discourse on existential risks (x-risks) from AI typically focuses on abrupt, dire events caused by advanced AI systems, particularly those that might achieve or surpass human-level intelligence. These events have severe consequences that either lead to human extinction or irreversibly cripple human civilization to a point beyond recovery. This decisive view, however, often neglects the serious possibility of AI x-risk manifesting gradually through an incremental series of smaller yet interconnected disruptions, crossing critical thresholds over time. This paper contrasts the conventional decisive AI x-risk hypothesis with what I call an accumulative AI x-risk hypothesis. While the former envisions an overt AI takeover pathway, characterized by scenarios like uncontrollable superintelligence, the latter suggests a different pathway to existential catastrophes. This involves a gradual accumulation of AI-induced threats such as severe vulnerabilities and systemic erosion of critical economic and political structures. The accumulative hypothesis suggests a boiling frog scenario where incremental AI risks slowly undermine systemic and societal resilience until a triggering event results in irreversible collapse. Through complex systems analysis, this paper examines the distinct assumptions differentiating these two hypotheses. It is then argued that the accumulative view can reconcile seemingly incompatible perspectives on AI risks. The implications of differentiating between the two types of pathway—the decisive and the accumulative—for the governance of AI as well as long-term AI safety are discussed. |
| Date | 2025-03-30 |
| Language | en |
| Short Title | Two types of AI existential risk |
| Library Catalog | Springer Link |
| URL | https://doi.org/10.1007/s11098-025-02301-3 |
| Accessed | 4/14/2025, 11:11:54 AM |
| Publication | Philosophical Studies |
| DOI | 10.1007/s11098-025-02301-3 |
| Journal Abbr | Philos Stud |
| ISSN | 1573-0883 |
| Date Added | 4/14/2025, 9:45:03 PM |
| Modified | 4/14/2025, 9:45:03 PM |
| Item Type | Journal Article |
|---|---|
| Author | Melanie Mitchell |
| Date | 2025-03-20 |
| Library Catalog | science.org (Atypon) |
| URL | https://www.science.org/doi/10.1126/science.adw5211 |
| Accessed | 3/25/2025, 5:54:35 PM |
| Extra | Publisher: American Association for the Advancement of Science |
| Volume | 387 |
| Pages | eadw5211 |
| Publication | Science |
| DOI | 10.1126/science.adw5211 |
| Issue | 6740 |
| Date Added | 4/2/2025, 2:13:04 PM |
| Modified | 4/2/2025, 2:13:04 PM |
| Item Type | Journal Article |
|---|---|
| Author | Sven Neth |
| Abstract | Hadfield-Menell et al. (2017) propose the Off-Switch Game, a model of Human-AI cooperation in which AI agents always defer to humans because they are uncertain about our preferences. I explain two reasons why AI agents might not defer. First, AI agents might not value learning. Second, even if AI agents value learning, they might not be certain to learn our actual preferences. |
| Date | 2025-02-26 |
| Language | en |
| Library Catalog | Springer Link |
| URL | https://doi.org/10.1007/s11098-025-02296-x |
| Accessed | 4/14/2025, 9:23:30 PM |
| Publication | Philosophical Studies |
| DOI | 10.1007/s11098-025-02296-x |
| Journal Abbr | Philos Stud |
| ISSN | 1573-0883 |
| Date Added | 4/14/2025, 9:45:03 PM |
| Modified | 4/14/2025, 9:45:03 PM |
| Item Type | Preprint |
|---|---|
| Author | Mikel Rodriguez |
| Author | Raluca Ada Popa |
| Author | Four Flynn |
| Author | Lihao Liang |
| Author | Allan Dafoe |
| Author | Anna Wang |
| Abstract | As frontier models become more capable, the community has attempted to evaluate their ability to enable cyberattacks. Performing a comprehensive evaluation and prioritizing defenses are crucial tasks in preparing for AGI safely. However, current cyber evaluation efforts are ad-hoc, with no systematic reasoning about the various phases of attacks, and do not provide a steer on how to use targeted defenses. In this work, we propose a novel approach to AI cyber capability evaluation that (1) examines the end-to-end attack chain, (2) helps to identify gaps in the evaluation of AI threats, and (3) helps defenders prioritize targeted mitigations and conduct AI-enabled adversary emulation to support red teaming. To achieve these goals, we propose adapting existing cyberattack chain frameworks to AI systems. We analyze over 12,000 instances of real-world attempts to use AI in cyberattacks catalogued by Google's Threat Intelligence Group. Using this analysis, we curate a representative collection of seven cyberattack chain archetypes and conduct a bottleneck analysis to identify areas of potential AI-driven cost disruption. Our evaluation benchmark consists of 50 new challenges spanning different phases of cyberattacks. Based on this, we devise targeted cybersecurity model evaluations, report on the potential for AI to amplify offensive cyber capabilities across specific attack phases, and conclude with recommendations on prioritizing defenses. In all, we consider this to be the most comprehensive AI cyber risk evaluation framework published so far. |
| Date | 2025-03-14 |
| Library Catalog | arXiv.org |
| URL | http://arxiv.org/abs/2503.11917 |
| Accessed | 3/25/2025, 5:59:00 PM |
| Extra | arXiv:2503.11917 [cs] |
| DOI | 10.48550/arXiv.2503.11917 |
| Repository | arXiv |
| Archive ID | arXiv:2503.11917 |
| Date Added | 4/2/2025, 2:13:04 PM |
| Modified | 4/2/2025, 2:13:04 PM |
| Item Type | Journal Article |
|---|---|
| Author | Rachel Etta Rudolph |
| Author | Elay Shech |
| Author | Michael Tamir |
| Abstract | Large language models (LLMs) such as OpenAI’s ChatGPT reflect, and can potentially perpetuate, social biases in language use. Conceptual engineering aims to revise our concepts to eliminate such bias. We show how machine learning and conceptual engineering can be fruitfully brought together to offer new insights to both conceptual engineers and LLM designers. Specifically, we suggest that LLMs can be used to detect and expose bias in the prototypes associated with concepts, and that LLM de-biasing can serve conceptual engineering projects that aim to revise such conceptual prototypes. At present, these de-biasing techniques primarily involve approaches requiring bespoke interventions based on choices of the algorithm’s designers. Thus, conceptual engineering through de-biasing will include making choices about what kind of normative training an LLM should receive, especially with respect to different notions of bias. This offers a new perspective on what conceptual engineering involves and how it can be implemented. And our conceptual engineering approach also offers insight, to those engaged in LLM de-biasing, into the normative distinctions that are needed for that work. |
| Date | 2025-02-18 |
| Language | en |
| Library Catalog | Springer Link |
| URL | https://doi.org/10.1007/s11098-024-02273-w |
| Accessed | 4/14/2025, 9:23:34 PM |
| Publication | Philosophical Studies |
| DOI | 10.1007/s11098-024-02273-w |
| Journal Abbr | Philos Stud |
| ISSN | 1573-0883 |
| Date Added | 4/14/2025, 9:45:03 PM |
| Modified | 4/14/2025, 9:45:03 PM |
| Item Type | Journal Article |
|---|---|
| Author | Rohin Shah |
| Author | Alex Irpan |
| Author | Alexander Matt Turner |
| Author | Anna Wang |
| Author | Arthur Conmy |
| Author | David Lindner |
| Author | Jonah Brown-Cohen |
| Author | Lewis Ho |
| Author | Neel Nanda |
| Author | Raluca Ada Popa |
| Author | Rishub Jain |
| Author | Rory Greig |
| Author | Scott Emmons |
| Author | Sebastian Farquhar |
| Author | Sébastien Krier |
| Author | Senthooran Rajamanoharan |
| Author | Sophie Bridgers |
| Author | Tobi Ijitoye |
| Author | Tom Everitt |
| Author | Victoria Krakovna |
| Author | Vikrant Varma |
| Author | Vladimir Mikulik |
| Author | Zachary Kenton |
| Author | Dave Orr |
| Author | Shane Legg |
| Author | Noah Goodman |
| Author | Allan Dafoe |
| Author | Four Flynn |
| Author | Anca Dragan |
| Language | en |
| Library Catalog | Zotero |
| Date Added | 4/14/2025, 9:44:59 PM |
| Modified | 4/14/2025, 9:44:59 PM |
| Item Type | Journal Article |
|---|---|
| Author | Z. Huey Wen |
| Abstract | The chemistry of combining the simulation hypothesis (which many believe to be a modern variation of skepticism) and manipulation arguments will be explored for the first time in this paper. I argue: If we take the possibility that we are now in a simulation seriously enough, then contrary to a common intuition, manipulation very likely does not undermine moral responsibility. To this goal, I first defend the structural isomorphism between simulation and manipulation: Provided such isomorphism, either both of them are compatible with moral responsibility, or none of them is. Later, I propose two kinds of reasons – i.e., the simulator-centric reason and the simulatee-centric reason – for why we have (genuine) moral responsibilities even if we are in a simulation. I close by addressing the significance of this paper in accounting for the relevance of artificial intelligence and its philosophy, in helping resolve a long-locked debate over free will, and in offering one reminder for moral responsibility specialists. |
| Date | 2025-03-17 |
| Language | en |
| Short Title | Simulation & Manipulation |
| Library Catalog | DOI.org (Crossref) |
| URL | https://www.cambridge.org/core/product/identifier/S1742360025000085/type/journal_article |
| Accessed | 3/17/2025, 9:41:12 PM |
| Pages | 1-16 |
| Publication | Episteme |
| DOI | 10.1017/epi.2025.8 |
| Journal Abbr | Episteme |
| ISSN | 1742-3600, 1750-0117 |
| Date Added | 4/2/2025, 2:09:25 PM |
| Modified | 4/2/2025, 2:09:25 PM |