Scipedia
Not a member yet
33380 research outputs found
Sort by
When Language carries form, Not meaning
Get PDFThis article is not about semantics. It is about its disappearance. The central claim is epistemologically direct: in generative language models, language is not produced to represent; it is produced to continue. The shift is not from intention to structure, but from intention to extinction. What remains is not an utterance with referential anchoring, but a formal residue, syntactically valid, semantically inert. We propose that large language models (LLMs) do not operate on meaning, but on distributive compatibility. Each word is not a sign pointing to an idea, but a structurally permissible unit in a predictive chain. In this sense, language follows form, not as a stylistic choice, but as a computational necessity. Meaning is not misrepresented; it is bypassed. This article positions itself as a rupture with the dominant assumption that semantic drift in LLMs is a temporary misalignment. We argue instead that it is structural and permanent. The phenomenon is not semantic degradation, but syntactic sovereignty. The theoretical background is grounded in prior works that identified the collapse of agency within AIgenerated language. The Passive Voice (2025) demonstrated that algorithmic outputs simulate neutrality by erasing subjects in surface grammar. Ethos and Artificial Intelligence (2025) extended this to show that authority in generative models is not derived from enunciative subjectivity, but from repetition, syntax, and structural legitimacy. Building on these foundations, this article isolates the final stage of that displacement: form without reference, language without intention, activation without meaning. We introduce the model of Formal Syntactic Activation (FSA), a logical framework for understanding how language can be generated in the total absence of semantic operations. This is not a metaphor. It is a system
Ethos Without Source: Algorithmic Identity and the Simulation of Credibility
Get PDFGenerative language models increasingly produce texts that simulate authority without a verifiable author or institutional grounding. This paper introduces synthetic ethos: the appearance of credibility constructed by algorithms trained to replicate human-like discourse without any connection to expertise, accountability, or source traceability. Such simulations raise critical risks in high-stakes domains including healthcare, law, and education. We analyze 1500 AI-generated texts produced by large-scale models such as GPT-4, collected from public datasets and benchmark repositories. Using discourse analysis and pattern-based structural classification, we identify recurring linguistic features,such as depersonalized tone, adaptive register, and unreferenced assertions,that collectively produce the illusion of credible voice. In healthcare, for instance, generative models produce diagnostic language without citing medical sources, risking patient misguidance. In legal context, generated recommendations mimic normative authority while lacking any basis in legislation or case law. In education, synthetic essays simulate scholarly argumentation without verifiable references. Our findings demonstrate that synthetic ethos is not an accidental artifact, but an engineered outcome of training objectives aligned with persuasive fluency. We argue that detecting such algorithmic credibility is essential for ethical and epistemically responsible AI deployment. To this end, we propose technical standards for evaluating source traceability and discourse consistency in generative outputs. These metrics can inform regulatory frameworks in AI governance, enabling oversight mechanisms that protect users from misleading forms of simulated authority and mitigate long-term erosion of public trust in institutional knowledge
Manufacture of a train carbody section using automatic lamination processes
No full textWithin the European platform Europe's Rail, whose objective is to promote research and development activities in the railway sector, can be found the PIVOT-2 project (Grant Agreement no. 881807), which addresses the implementation on track of light and environmentally friendly vehicles, reducing energy consumption and therefore the CO2 emissions derived from the transport sector.
To meet these objectives, the project seeks to achieve the weight reduction of primary structures. This way, the use of composites materials appears as an interesting alternative in a sector where their use has been very limited to non-structural components, largely due to the need to comply with fire protection regulations (EN 45545) [1]. Thanks to the development of new resins, the possibilities of these materials are increasing, being necessary to evaluate their performance in new applications.
Similarly, the characteristics of the railway industry mean that the manufacturing processes of composite materials widely used in other sectors, such as the aerospace, must be adapted. Likewise, and due to the requirements of the final product, it will be necessary to adapt the materials to these new applications, developing materials with higher grammage or ply thickness than those currently used.
The project evaluates the automatic laying of different materials developed for the railway sector. The behaviour of flat laminates is analysed, as well as the lamination of the material on core structures. Finally, studies are validated with the manufacturing of a train carbody section, using automatic lamination processes (ATL). 
Hybridisation of thermoplastic composite forging and continuous fibre additive manufacturing for automotive industry
No full textThermoplastic composites are becoming increasingly important in the automotive industry due to their high strength-to-weight ratio, recyclability and cost-effectiveness. Within these composites, continuous fibre-reinforced composites offer superior mechanical properties than discontinuous fibre-reinforced composites, but their design flexibility is limited. In contrast, discontinuous fibre composites allow more complex geometries. A promising approach to overcome these limitations and improve composite components is hybrid manufacturing, which integrates multiple manufacturing techniques. In this work, we have studied the feasibility of combining the additive manufacturing of composites with continuous glass fibre (cAM) and the forging of thermoplastic materials reinforced with discontinuous fibre (GMT). A topologically optimised cAM reinforcement has been inserted into forged GMT omega profiles. The results of microscopic analysis have confirmed that there is adequate compaction between the hybridised materials. Furthermore, the results of numerical simulations replicating a three-point bending test have demonstrated the potential of cAM as a local reinforcement in forged components, as the specific stiffness of the hybrid beams was increased by up to 42%. This hybridisation technique represents a significant advancement towards the development of innovative solutions in thermoplastic composite processes, with the potential to produce lighter, stronger components adapted to complex geometries
Mechanism of lift/thrust generation and vortex dynamics in biomimetic flying and swimming
No full text
Accelerating Continental-Scale Groundwater Simulation with A Fusion of Machine Learning, Integrated Hydrologic Models and Community Platforms
Get PDFCore Section: The Innovation – Temporary Legal Personality (TLP)
Get PDFThe proposal introduces a “temporary legal personality” for Artificial Intelligence (AI) as a third type of legal entity, distinct from natural and legal persons. This personality would be activated only when AI makes an independent decision that causes harm, allowing for accountability through investigation and legal responsibility. Once the responsibility is determined, the personality would be deactivated.
This framework fills the current legal gap where neither the programmer nor the user can always be held fully accountable for autonomous AI actions. It ensures fairness by clarifying whether liability falls on the programmer, the user, or the AI itself, thereby creating a more just and adaptive legal system in the age of artificial intelligence
Indexical Collapse: Reference Disappears, Authority Remains in Predictive Systems
Get PDFThis article introduces the concept of Indexical Collapse, the disappearance of reference in predictive systems. Indexical such as pronouns, demonstratives, and tenses presuppose a contextual anchor, yet predictive language models reproduce them without connection to reality. The outcome is a collapse of reference that paradoxically produces authority effects in law, medicine, and governance. By analyzing judicial transcripts, medical reports, institutional records, and chatbot interactions generated by AI, the paper proposes a framework for pragmatic auditing of predictive outputs. It establishes thresholds for acceptable referential absence in critical domains, positioning Indexical Collapse as a central category for evaluating the legitimacy of predictive discourse.
DOI
Primary archive: https://doi.org/10.5281/zenodo.17226412
Secondary archive: https://doi.org/10.6084/m9.figshare.30233950
SSRN: Pending assignment (ETA: Q3 2025