Human Centred AI in the post optimization regenerative process for climate scenarios. Adressing new workflow for Re-Generative Digital Design

Does the use of AI in design experiments only mean using DL (deep learning) to inform CV (computer vision) with a huge amount of data in order to return a desired image? Can AI be used not only in applications of environmental disciplines, visioning of future cities and text-image generation, but also in other pre-visualization processes? This paper is a contribution to the ongoing discussion on the scientific question and emerging technologies, whether it is possible to address the use of AI-assisted design (AIAD), with human centered approach, to the processes of parametric de-sign integrating climatic scenarios, in the digital scale jumping phase for under-star-ting conflicting objectives; to address multi-domain integration through AI-assisted deep learning, to algorithms that use ‘prompts as natural language processing (NLP)’ to guide the choice of response alternatives, to overcome errors in the optimization of climate data (discretization processes) and to couple the simulation of predictive models in the tools’ own digital language, images and message-carrying configurations to be used in the pre-design phase. Therefore, it is necessary to formulate a new background, rather than a literature, with a “scoping review” on different scientific fields, in order to trace the trajectories of new paradigms, thus the relations and cross-references between the terms and the repercussions on interdisciplinary issues of technological and environmental design. From a methodological point of view, in order to formulate instructions for a new workflow for re-generative digital design, in the pre-design phase under climatic scenarios, the operational steps that characterize parametric digital processes for the formulation of predictive models are taken to hybridize them with steps that are more typical of the algorithmic processes of AI, where the machine learning code is hybridized with the information of deep algorithmic learning, for the integrated visualization of solutions to be discussed in environmental and technological terms. The conclusions are entrusted with some perspective reflections, useful to build theoretical apparatuses on the “responsibility of innovation” and its application frameworks referable to frontier research in climate change design, through human-centred AI.