Epistemoverse: Toward an AI-Driven Knowledge Metaverse for Intellectual Heritage Preservation

Epistemoverse: Toward an AI-Driven Knowledge Metaverse for Intellectual Heritage Preservation Predrag K. Nikolić School of Design and Arts Swinburne University of Technology Kuching, Malaysia pnikolic@swin.edu.au Robert Prentner Institute of Humanities ShanghaiTech University Shanghai, China robert.prentner@amcs.science Abstract Large language models (LLMs) have often been characterized as “stochastic parrots” that merely reproduce fragments of their train- ing data. This study challenges that assumption by demonstrating that, when placed in an appropriate dialogical context, LLMs can de- velop emergent conceptual structures and exhibit interaction-driven (re-)structuring of cognitive interfaces and reflective question-asking. Drawing on the biological principle of cloning and Socrates’ maieu- tic method, we analyze authentic philosophical debates generated among AI-reincarnated philosophers within the interactive art in- stallations of the Syntropic Counterpoints project. By engaging digital counterparts of Aristotle, Nietzsche, Machi- avelli, and Sun Tzu in iterative discourse, the study reveals how machine dialogue can give rise to inferential coherence, reflective questioning, and creative synthesis. Based on these findings, we pro- pose the concept of the Epistemoverse—a metaverse of knowledge where human and machine cognition intersect to preserve, reinter- pret, and extend intellectual heritage through AI-driven interaction. This framework positions virtual and immersive environments as new spaces for epistemic exchange, digital heritage, and collabora- tive creativity. CCS Concepts • Applied computing →Arts and humanities; • Human-centered computing →Interaction design; • Hardware →Emerging technologies; • Mathematics of computing →Graph theory. ACM Reference Format: Predrag K. Nikolić and Robert Prentner. 2025. Epistemoverse: Toward an AI- Driven Knowledge Metaverse for Intellectual Heritage Preservation. In The 20th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry (VRCAI ’25), December 13–14, 2025, Macau, China. ACM, New York, NY, USA, 7 pages. https://doi.org/10.1145/3779232. 3779278 1 Introduction In biology, a clone is defined as an individual that is genetically identical to the original organism from which the clone was derived Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from permissions@acm.org. VRCAI ’25, Macau, China © 2025 Copyright held by the owner/author(s). Publication rights licensed to ACM. ACM ISBN 979-8-4007-2362-9/2025/12 https://doi.org/10.1145/3779232.3779278 [Raff 2018]. Throughout biology, cloning is a ubiquitous reproduc- tive strategy that can be observed in a variety of species, mostly in bacteria and plants. It is essential to emphasize that a clone dif- fers from a copy. Whereas a clone is genetically identical to the organism it is derived from, it can potentially undergo an individ- ual developmental trajectory. Virtually all biological traits, such as cognitive differences among individuals, arise as an intricate com- bination of genetic predisposition and environmental facilitation [Bouchard Jr. and McGue 2003]. This is different from the way we typically think about software agents, which are mere copies of a program instantiated on different computers and executed in a well-defined and identical manner. In this contribution, we are investigating AI clones of historically existing philosophers [Nikolić et al. 2021a]. We aim to demonstrate that those clones, while implemented via fine-tuning an LLM [Fierro et al. 2024] on the corpus of their “ancestors” (Aristotle, Nietzsche, Machiavelli, and Sun Tzu, respectively), exhibit remarkable kinds of knowledge that go beyond the knowledge that we would have uncovered if we studied them in isolation. Analysis should focus on the intrinsic perspective of philoso- phers, not on the correspondence between their utterances and hu- man knowledge. We are not interested in how those clones would fare against human philosophers or whether they are just repeat- ing what those philosophers have said (unlike what was presented in recent lawsuits against AI companies, who allegedly violated the copyrights of real human authors). By contrast, we are mainly interested in whether our AI clones could produce knowledge that stems from interaction-induced conceptual organization, beyond isolated reproduction. Put more provocatively, their knowledge might be said to be “genuinely theirs.

This content is AI-processed based on open access ArXiv data.