I have been experimenting with practical ways to use AI in local planning and community documentation. This may be useful to people here who follow Jacksonville development, transit, architecture, or neighborhood history.
The best results come from treating AI as a visual sketchbook, not a replacement for photography or professional design. Start with a clear reference image, preserve the original, and change one variable at a time. A streetscape photo can illustrate how a pedestrian plaza, crosswalk treatment, or shade trees might change a block. It is not a construction drawing, but it can make comparisons easier.
Pose and viewpoint also matter when documenting events or testing a public-space layout. A consistent angle makes before-and-after comparisons more credible. I found an AI pose changer (https://imgtoimg-ai.com/ai-image-editor/pose-changer-ai) useful for controlled variations. I label generated images and keep the original beside them so nobody mistakes an illustration for a record.
My workflow is simple: save the original with date and location; crop before editing; change one variable per version; use descriptive filenames; compare the edit with the source; and share prompts when an image is used publicly.
Jacksonville's bright sun, deep shadows, and active construction can make a composite look more certain than its evidence. Reflections, signs, lane markings, and tree canopies are easy for an image model to alter accidentally. For neighborhood history, accuracy should come first and creative edits should be clearly marked. Showing several alternatives is better than presenting one preferred outcome.
Used carefully, these tools can help residents explain an idea and invite feedback without pretending that a generated picture is a survey or approved plan. What workflows do others use for mapping, accessibility audits, transit advocacy, or documenting changes around town?