<p>We investigated the potential of applying artificial intelligence (AI) to assist in the landscape design of green spaces that consider not only their physical attributes but also the user’s impressions. In a questionnaire, seven participants were asked for their impressions of 23 images of green spaces, and the results were used to analyze the relationship between their impressions and the physical elements of the green spaces. A psychological evaluation framework was established for understanding how the physical elements of green spaces influence user impressions. Based on this framework, we used AI image editing functions in Adobe Photoshop to modify the original images and incorporate physical elements expected to improve the user’s impression of the green space, such as water features and seasonal changes. We then conducted a second questionnaire to evaluate whether the edited images led to any changes in impressions. The results suggested that the added physical elements may contribute to enhancing user impressions of green spaces. These findings suggest that the proposed psychological evaluation framework has the potential to support AI-assisted landscape design processes that are more closely aligned with human sensibilities. This study provides fundamental insights for the future development of urban green spaces and AI-assisted design tools. As a pilot study, this research provides preliminary insights into the integration of psychological evaluation and AI-assisted design support for green spaces.</p>