This positional paper shows the use of video generation models, focusing on prompt engineering to guide the generation of video prototyping. Using generative AI (GenAI), designers and researchers can accelerate immersive prototyping, enabling rapid exploration and iteration of interaction designs and integrating GenAI tools that facilitate visualization into the interaction design process. Different input prompt alternatives are compared across video generation models to create prototypes representing the desired interactions in immersive environments, providing user-specific text prompts. Along with a comparative analysis of these technologies, two text prompts are used as input to three different text-to-video (T2V) generation models evaluating the capacity to create video prototyping based on text prompts. The evaluation was conducted with five industry experts in computer animation, artificial intelligence, and immersive systems who assessed the video quality, animation fluency, and message coherence for each video prototype generated. The results revealed significant differences in the perception of videos generated by different models and prompts. Additionally, the limitations and opportunities of video prototyping in immersive systems are discussed, highlighting the importance of considering visual quality, narrative, and user experience in the design process. Through this investigation, the research aims to provide insights into the efficacy of prompt-driven video prototyping and its implications for immersive system design.