Toward Agentic Environments: <scp>GenAI</scp> and the Convergence of <scp>AI</scp> , Sustainability, and Human‐Centric Spaces

ABSTRACT In the past few years, the evolution of artificial intelligence (AI), particularly generative AI (GenAI) and large language models (LLMs), has made human‐computer interactions more frequent, easier, and faster than ever before. This brings numerous benefits in terms of enhancing efficiency, accessibility, and convenience in various sectors from banking to health. AI tools and solutions applied in computers and communication devices support decision‐making processes and managing operations of users on the individual level as well as organisationally, including resource allocation, workflow automation, and real‐time data analysis. However, the current use of AI carries a substantial environmental footprint due to its reliance on high‐computational cloud resources. In such a context, this paper introduces the concept of agentic environments, a sustainability‐oriented AI framework that goes beyond reactive systems by leveraging GenAI, multi‐agent systems, and edge computing to minimize the negative impact of technology. These types of environments can contribute to the optimization of resource use, enhanced quality of life, and prioritization of sustainability while at the same time safeguarding user privacy through decentralized, edge‐driven AI solutions. Based on both secondary and primary data gathered during a focus group and semi‐structured interviews with AI professionals from leading technology companies, the authors provide a conceptual framework of agentic environments and discuss it in the context of three lenses, including personal sphere, professional and commercial use, and urban operations. The findings include the potential of agentic environments to foster sustainable ecosystems, mainly due to the optimisation of resource usage and securing the privacy of data. The study outlines recommendations for implementing edge‐driven deployment models to reduce dependency on currently widely applied high‐energy cloud solutions.