This thesis explores the use of conditional Generative Adversarial Networks (cGANs) for a collaborative, designer-informed workflow in the context of Additive Manufacturing (AM). It addresses the challenges of enabling designer interaction and control over AI-generated outcomes, aiming to generate a more responsive design-to-fabrication process. To achieve this, the system captures designer-guided modifications and design forms and translates 3D modeling information and fabrication parameters into a 2D pixel-based representation, forming customized datasets for machine learning training. The model then learns the logic of specific design intentions and fabrication modification styles through training, which informs future users and supports adaptive design interpretation. This study investigates how designers and AI collaborate in generative design applications through material and fabrication experiments. The results show the potential of a human-in-the-loop fabrication system, where AI informs the designers with adaptive design decisions and contributes to fabrication-aware design processes. This research expands the possibilities of digital craftsmanship by connecting computational intelligence with intuitive design intent.