{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T15:00:54Z","timestamp":1773932454015,"version":"3.50.1"},"reference-count":48,"publisher":"MDPI AG","issue":"4","license":[{"start":{"date-parts":[[2024,12,15]],"date-time":"2024-12-15T00:00:00Z","timestamp":1734220800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["MAKE"],"abstract":"<jats:p>Large language models (LLMs) achieve remarkable predictive capabilities but remain opaque in their internal reasoning, creating a pressing need for more interpretable artificial intelligence. Here, we propose bridging this explanatory gap by drawing on concepts from topological quantum computing (TQC), specifically the anyonic frameworks arising from SU(2)k theories. Anyons interpolate between fermions and bosons, offering a mathematical language that may illuminate the latent structure and decision-making processes within LLMs. By examining how these topological constructs relate to token interactions and contextual dependencies in neural architectures, we aim to provide a fresh perspective on how meaning and coherence emerge. After eliciting insights from ChatGPT and exploring low-level cases of SU(2)k models, we argue that the machinery of modular tensor categories and topological phases could inform more transparent, stable, and robust AI systems. This interdisciplinary approach suggests that quantum-theoretic principles may underpin a novel understanding of explainable AI.<\/jats:p>","DOI":"10.3390\/make6040137","type":"journal-article","created":{"date-parts":[[2024,12,17]],"date-time":"2024-12-17T06:26:12Z","timestamp":1734416772000},"page":"2876-2891","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":5,"title":["What ChatGPT Has to Say About Its Topological Structure: The Anyon Hypothesis"],"prefix":"10.3390","volume":"6","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-5739-546X","authenticated-orcid":false,"given":"Michel","family":"Planat","sequence":"first","affiliation":[{"name":"Institut FEMTO-ST CNRS UMR 6174, Universit\u00e9 de Franche-Comt\u00e9, 15 B Avenue des Montboucons, F-25044 Besan\u00e7on, France"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0637-1916","authenticated-orcid":false,"given":"Marcelo","family":"Amaral","sequence":"additional","affiliation":[{"name":"Quantum Gravity Research, Los Angeles, CA 90290, USA"}]}],"member":"1968","published-online":{"date-parts":[[2024,12,15]]},"reference":[{"key":"ref_1","unstructured":"(2024, October 01). Pause Giant AI Experiments: An Open Letter. Available online: https:\/\/en.wikipedia.org\/wiki\/Pause_Giant_AI_Experiments:An_Open_Letter."},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Bryant, A. (2023). AI chatbots: Threat or opportunity?. Informatics, 10.","DOI":"10.3390\/informatics10020049"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"e1391","DOI":"10.1002\/widm.1391","article-title":"A historical perspective of explainable artificial intelligence","volume":"11","author":"Confalioneri","year":"2019","journal-title":"Data Mining Know. Discov."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"5031","DOI":"10.1109\/TII.2022.3146552","article-title":"From artificial intelligence to explainable artificial intelligence in industry 4.0: A survey on what, how, and where","volume":"18","author":"Ahmed","year":"2022","journal-title":"IEEE Trans. Indust. Inform."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"615","DOI":"10.3390\/make3030032","article-title":"Classification of explainable artificial intelligence methods through their output formats","volume":"3","author":"Vilone","year":"2021","journal-title":"Mach. Learn. Knowl. Extr."},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Ferraz-Caetano, J. (2023). The artificial intelligence explanatory trade-Off on the logic of discovery in chemistry. Philosophies, 8.","DOI":"10.3390\/philosophies8020017"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Tahir, H.A., Alayed, W., Hassan, W.U., and Haider, A. (2024). Proposed explainable interference control technique in 6G networks using large language models (LLMs). Electronics, 13.","DOI":"10.3390\/electronics13224375"},{"key":"ref_8","first-page":"115","article-title":"Why Topology for machine learning and knowledge extraction?","volume":"1","author":"Ferri","year":"2019","journal-title":"Mach. Learn. Extr."},{"key":"ref_9","first-page":"1","article-title":"Introduction to Machine Learning & Knowledge Extraction (MAKE)","volume":"1","author":"Holzinger","year":"2019","journal-title":"Mach. Learn. Extr."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Feng, J.H., Lai, E.M.-K., and Li, W. (2024). Geometry of textual data augmentation: Insights from large language models. Electronics, 13.","DOI":"10.3390\/electronics13183781"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Lozi\u0107, E., and \u0160tular, B. (2023). Fluent but not factual: A comparative analysis of ChatGPT and other AI chatbots\u2019 proficiency and originality in scientific writing for humanities. Future Internet, 15.","DOI":"10.3390\/fi15100336"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"219","DOI":"10.4236\/jcc.2024.123014","article-title":"Evaluating ChatGPT\u2019s consciousness and its capability to pass the Turing test: A comprehensive analysis","volume":"12","author":"Gams","year":"2024","journal-title":"J. Comp. Comm."},{"key":"ref_13","unstructured":"Chen, S., Yu, S., Zhao, S., and Lu, C. (2024). From imitation to introspection: Probing self-consciousness in language models. arXiv."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Pira, L., and Ferrie, C. (2024). On the interpretability of quantum neural networks. arXiv.","DOI":"10.1007\/s42484-024-00191-y"},{"key":"ref_15","unstructured":"Steinm\u00fcller, P., Schulz, T., Graf, F., and Herr, D. (2022). eXplainable AI for quantum machine learning. arXiv."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"164","DOI":"10.1016\/j.neunet.2022.05.028","article-title":"Quantum neural networks and topological quantum field theories","volume":"153","author":"Marciano","year":"2022","journal-title":"Neural Netw."},{"key":"ref_17","unstructured":"Tommaso, T. (2024, November 15). Interpretability and Explainability AI in Advanced Neural Networks for Topological Quantum Field Theory and Medical Application. Available online: https:\/\/inspirehep.net\/literature\/2849341."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"045004","DOI":"10.1088\/2058-9565\/aacad2","article-title":"Introduction to topological quantum computation with non-Abelian anyons","volume":"3","author":"Field","year":"2018","journal-title":"Quantum Sci. Technol."},{"key":"ref_19","unstructured":"Wang, Z. (2010). Topological Quantum Computing, American mathematical Society."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"1083","DOI":"10.1103\/RevModPhys.80.1083","article-title":"Non-Abelian anyons and topological quantum computation","volume":"80","author":"Nayak","year":"2008","journal-title":"Rev. Mod. Phys."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Biedenharn, L.C., and Lohe, M.A. (1995). Quantum Group Symmetry and q-Tensor Algebras, World Scientific.","DOI":"10.1142\/2815"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"195129","DOI":"10.1103\/PhysRevB.107.195129","article-title":"Numerical simulation of non-abelian anyons","volume":"107","author":"Kirchner","year":"2023","journal-title":"Phys. Rev. B"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"060301","DOI":"10.1088\/0256-307X\/40\/6\/060301","article-title":"Digital simulation of non-Abelian anyons with 68 programmable superconducting qubits","volume":"40","author":"Xu","year":"2023","journal-title":"Chinese Phys. Lett."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"505","DOI":"10.1038\/s41586-023-06934-4","article-title":"Non-abelian topological order and anyons on a trapped-ion processor","volume":"626","author":"Iqbal","year":"2024","journal-title":"Nature"},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"343","DOI":"10.1007\/s00220-009-0908-z","article-title":"On the classification of modular tensor categories","volume":"292","author":"Rowell","year":"2009","journal-title":"Comm. Math. Phys."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Planat, M., and Chester, D. (2024). Topology and dynamics of transcriptome (dys)regulation. Int. J. Mol. Sci., 25.","DOI":"10.20944\/preprints202403.1741.v1"},{"key":"ref_27","unstructured":"Knight, W. (2024, November 11). OpenAI\u2019s CEO Says the Age of Giant AI Models Is Already Over. Wired. Available online: https:\/\/www.wired.com\/story\/openai-ceo-sam-altman-the-age-of-giant-ai-models-is-already-over\/."},{"key":"ref_28","unstructured":"Villalobos, P., Ho, A., Sevilla, J., Besiroglu, T., Heim, L., and Hobbhahn, M. (2024). Will we run out of data? Limits of LLM scaling based on human-generated data. arXiv."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"360","DOI":"10.1016\/0550-3213(88)90603-7","article-title":"Fusion rules and modular transformations in 2D conformal theory","volume":"B300","author":"Verlinde","year":"1988","journal-title":"Nucl. Phys."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"235120","DOI":"10.1103\/PhysRevB.87.235120","article-title":"Anyonic quantum spin chains: Spin 1 generalizations and topological stability","volume":"87","author":"Gils","year":"2008","journal-title":"Phys. Rev. B"},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Kac, V.G. (1988). Representations of the algebra Uq(sl(2)), q-orthogonal polynomials and invariants of links. Infinite Dimensional Lie Algebras and Groups, World Scientific.","DOI":"10.1142\/9789812798343"},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"010334","DOI":"10.1103\/PRXQuantum.2.010334","article-title":"Fibonacci anyons versus Majorana fermions: A Monte Carlo approach to the compilation of braid circuits in SU(2)k anyon models","volume":"2","author":"Johansen","year":"2021","journal-title":"PRX Quantum"},{"key":"ref_33","unstructured":"Aboumrad, W. (2022). Quantum computing with anyons: An F-matrix and braid calculator. arXiv."},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Amaral, M., Chester, D., Fang, F., and Irwin, K. (2022). Exploiting anyonic behavior of quacicrystals for topological qiantum computing. Symmetry, 14.","DOI":"10.3390\/sym14091780"},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"2509","DOI":"10.1007\/s11128-013-0544-6","article-title":"A new set of generators and a physical interpretation for the SU(3) finite subgroup D(9,1,1;2,1,1)","volume":"7","author":"Bauer","year":"2013","journal-title":"Quantum Inf. Proc."},{"key":"ref_36","unstructured":"Levaillant, C. (2013). On some projective unitary qutrit gates. arXiv."},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Planat, M., Aschheim, R., Amaral, M.M., and Irwin, K. (2020). Informationally complete characters from quark and lepton mixings. Symmetry, 12.","DOI":"10.3390\/sym12061000"},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"335","DOI":"10.1038\/s41583-020-0277-3","article-title":"Back propagation and the brain","volume":"21","author":"Lillicrap","year":"2020","journal-title":"Nat. Rev. Neurosci."},{"key":"ref_39","unstructured":"Hofer, C., Kwitt, R., Niethammer, M., and Uhl, A. (2017). Deep Learning with Topological Signatures. Advances in Neural Information Processing Systems (NeurIPS) 30, Neural Information Processing Systems Foundation, Inc."},{"key":"ref_40","unstructured":"Zheng, Z., Wang, Y., Huang, Y., Song, S., Yang, M., Tang, B., Xiong, F., and Li, Z. (2024). Attention heads of large language models: A survey. arXiv."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"3713","DOI":"10.1007\/s11042-022-13428-4","article-title":"Natural language processing: State of the art, curren trends and challenges","volume":"82","author":"Khurana","year":"2022","journal-title":"Multimed. Tools Appl."},{"key":"ref_42","first-page":"604","article-title":"A survey of usages of deep learning for natural language processing","volume":"32","author":"Otter","year":"2021","journal-title":"IEEE Trans. Neural Netw."},{"key":"ref_43","doi-asserted-by":"crossref","unstructured":"Raynal, M. (2013). Concurrent Programming: Algorithms, Principles and Foundations, Springer.","DOI":"10.1007\/978-3-642-32027-9"},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"1191","DOI":"10.1007\/s00220-022-04517-4","article-title":"From three dimensional manifolds to modular tensor categories","volume":"397","author":"Cui","year":"2023","journal-title":"Comm. Math. Phys."},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"167","DOI":"10.1112\/S0024611504015011","article-title":"From Klein to Painlev\u00e9 via Fourier, Laplace and Jimbo","volume":"90","author":"Boalch","year":"2005","journal-title":"Proc. Lond. Math. Soc."},{"key":"ref_46","doi-asserted-by":"crossref","first-page":"1","DOI":"10.3390\/dynamics4010001","article-title":"Dynamics of Fricke-Painlev\u00e9 surfaces","volume":"4","author":"Planat","year":"2024","journal-title":"Dynamics"},{"key":"ref_47","doi-asserted-by":"crossref","first-page":"195145","DOI":"10.1103\/PhysRevB.96.195145","article-title":"Machine learning topological states","volume":"96","author":"Deng","year":"2017","journal-title":"Phys. Rev. B"},{"key":"ref_48","unstructured":"Sanborn, S., Mathe, J., Papillon, M., Buracas, D., Lillemark, H.J., Showmake, C., Bertics, A., Pennec, X., and Miolane, N. (2024). Beyond Euclid: An illustrated guide to modern machine learning with geometric, topological, and algebraic structures. arXiv."}],"container-title":["Machine Learning and Knowledge Extraction"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2504-4990\/6\/4\/137\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T16:52:34Z","timestamp":1760115154000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2504-4990\/6\/4\/137"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,12,15]]},"references-count":48,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2024,12]]}},"alternative-id":["make6040137"],"URL":"https:\/\/doi.org\/10.3390\/make6040137","relation":{},"ISSN":["2504-4990"],"issn-type":[{"value":"2504-4990","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,12,15]]}}}