TY - GEN
T1 - Regulatory Challenges of LLM-Integrated Blockchain Agents in the Metaverse
AU - Otalora, Francisco Cordoba
AU - Themistocleous, Marinos
N1 - Publisher Copyright:
© 2025 IEEE.
PY - 2025
Y1 - 2025
N2 - The metaverse, a virtual ecosystem, integrates large language models (LLMs) and blockchain agents to create immersive, decentralized interactions, raising complex regulatory challenges. This paper examines the governance hurdles of LLM-powered blockchain agents in the metaverse, focusing on data protection, accountability, and decentralized governance. Adopting a systematic analysis drawing on technical and legal perspectives, it explores how LLMs’ generative autonomy and blockchain’s immutability complicate compliance with privacy laws, liability frameworks, and ethical standards. Key findings reveal significant gaps: blockchain’s permanence conflicts with data protection rights like erasure, autonomous agents obscure accountability pathways, and decentralized systems often lack robust oversight, creating risks of bias and vulnerability. Proposed governance principles leveraging decentralized technologies like blockchain and DAOs alongside adaptive standards and participatory models offer potential solutions to enhance trust and fairness. The study underscores the urgent need for global coordination to harmonize regulations and develop infrastructure addressing these unique convergence challenges. Future research should explore areas such as the implications of asset tokenization and the development of scalable privacy frameworks to ensure equitable metaverse ecosystems. This work contributes to shaping a regulatory paradigm that balances innovation with ethical and legal integrity.
AB - The metaverse, a virtual ecosystem, integrates large language models (LLMs) and blockchain agents to create immersive, decentralized interactions, raising complex regulatory challenges. This paper examines the governance hurdles of LLM-powered blockchain agents in the metaverse, focusing on data protection, accountability, and decentralized governance. Adopting a systematic analysis drawing on technical and legal perspectives, it explores how LLMs’ generative autonomy and blockchain’s immutability complicate compliance with privacy laws, liability frameworks, and ethical standards. Key findings reveal significant gaps: blockchain’s permanence conflicts with data protection rights like erasure, autonomous agents obscure accountability pathways, and decentralized systems often lack robust oversight, creating risks of bias and vulnerability. Proposed governance principles leveraging decentralized technologies like blockchain and DAOs alongside adaptive standards and participatory models offer potential solutions to enhance trust and fairness. The study underscores the urgent need for global coordination to harmonize regulations and develop infrastructure addressing these unique convergence challenges. Future research should explore areas such as the implications of asset tokenization and the development of scalable privacy frameworks to ensure equitable metaverse ecosystems. This work contributes to shaping a regulatory paradigm that balances innovation with ethical and legal integrity.
KW - AI agent
KW - asset tokenization
KW - blockchain
KW - Large language model
KW - metaverse
UR - https://www.scopus.com/pages/publications/105032100852
U2 - 10.1109/iMETA66706.2025.11306431
DO - 10.1109/iMETA66706.2025.11306431
M3 - Conference contribution
AN - SCOPUS:105032100852
T3 - 3rd International Conference on Intelligent Metaverse Technologies and Applications, iMETA 2025
SP - 33
EP - 38
BT - 3rd International Conference on Intelligent Metaverse Technologies and Applications, iMETA 2025
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 3rd International Conference on Intelligent Metaverse Technologies and Applications, iMETA 2025
Y2 - 14 October 2025 through 17 October 2025
ER -