arXiv:2607. 25620v1 Announce Type: new Abstract: Quattrociocchi and colleagues warn that the fluent outputs of large language models may allow linguistic plausibility to substitute for epistemic evaluation, producing the condition they call *Epistemia*: the experience of possessing knowledge without undertaking the practices through which judgment would ordinarily be warranted.
The paper argues that new interactional concepts are needed to understand how humans change when engaging with large language models (LLMs). Drawing on enactive theory and the notion of Linguistic Bodies, it proposes that humans incorporate LLMs both as tools that enable specific skills and as linguistic partners that shape identity and language use. The authors introduce the idea of "susceptive sense‑making" to describe the heightened semiotic labor involved when friction is absent in these interactions, and suggest that this dynamic may affect users’ autonomy and development.
The article discusses the evolution of AI across three generations—from explicit logic to neural networks to large language models (LLMs)—and how LLMs introduce new systemic risks. It applies a forensic‑psychology profiling method to identify ten key features of LLMs, such as hallucinations, bias, and cognitive atrophy, revealing an entity that confabulates, amplifies user biases, and erodes human competence. The report concludes with a four‑pillar framework for AI resilience, emphasizing cognitive sovereignty, measurable control, partial autonomy, and openness to safeguard society.
arXiv:2608. 08443v1 Announce Type: cross Abstract: Previous studies have shown that people can develop shared symbols, partner-specific expressions, personal idioms, inside jokes, and other parts of a relational microculture.
The paper proposes a new interdisciplinary field called Cognitive Infrastructure Studies (CIS) to examine how AI systems act as invisible, foundational cognitive infrastructures that shape what people can know and do in digital societies. It argues that these infrastructures, through anticipatory personalization and adaptive invisibility, automate relevance judgments and shift epistemic agency to non‑human systems. CIS offers methodological tools, such as infrastructure breakdown experiments, to uncover the hidden cognitive dependencies created by AI preprocessing across individual, collective, and societal levels.
arXiv:2607. 01248v1 Announce Type: cross Abstract: Large language models are increasingly used for knowledge acquisition, code generation, academic writing, and agent-based automation.
The paper proposes a diagnostic analogy framework to analyze how generative AI reshapes cultural and epistemic practices. It breaks down interventions into three coordinates—epistemic site, governing logic, and technical mechanism—to differentiate structural consequences from those that can be designed. Applying this framework to information discovery and knowledge synthesis, the authors illustrate how shifts from indexicality to inference and from editorial authority to statistical consensus generate specific cultural effects and uncover governance levers hidden by broad analogies.
arXiv:1912. 08786v3 Announce Type: replace-cross Abstract: Three generations of software have transformed the role of artificial intelligence in society.
The article presents a minimal working model for large language model (LLM) systems, emphasizing four key distinctions—pretraining vs. deployment, distribution vs. samples, types of memory, and task competence vs. agency. Using this framework, it diagnoses six common misconceptions about LLMs (next‑token prediction, regression to the mean, training‑data regurgitation, model memory, alignment, and understanding), explaining what each misconception captures correctly, where it conflates distinctions, and the implications for evaluation, design, and governance. The model is applied to AI policy language, illustrating how policy can misrepresent these distinctions and offering a diagnostic toolkit to correct such errors.
The paper investigates when an interpretation in generative AI is considered established, arguing that passing local factual checks is insufficient. It introduces three concepts—interpretive appearance, evaluation contract, and standing substitution—to analyze how interpretations gain recognition within sociotechnical processes. The authors propose delayed closure as a practice to keep recognized interpretations revisable and outline five public requirements for transparency, evidence, failure handling, contract revision, and responsibility.
arXiv:2606. 12713v1 Announce Type: new Abstract: Claims that artificial general intelligence has already arrived and claims that it remains decades away are often defended from overlapping evidence.
The paper examines how Large Language Models (LLMs) and generative AI (GenAI) are biased against less-spoken languages and deepen the digital language divide. Drawing on critical sociolinguistics and computational linguistics, it investigates the intersections of technology, language policy, and linguistic variation using South Tyrolean dialects and Kurdish varieties as case studies. The authors explore technical solutions for handling non‑standard language and assess whether such approaches can support democratic and decolonial digital strategies with policy implications.