# How to Make AI Videos for Beginners: INSANE Results From 3 Prompts **URL:** \ **Category:** Content Creators\ **Tags:** prompt-engineer-48, no-priors:-ai, education, video, tech\ **Created:** [31 October 2026 03:46 UTC](https://www.artofsm.art/t/how-to-make-ai-videos-for-beginners-insane-results-from-3-prompts/27212 "2026-10-31T03:46:47Z")\ **Posts on this page:** 2\ **Page:** 1
**Author:** ![merefield](https://www.artofsm.art/user_avatar/www.artofsm.art/merefield/32/74_2.png) [@merefield](https://www.artofsm.art/u/merefield)\ **Post date:** [31 October 2026 03:46 UTC](https://www.artofsm.art/t/how-to-make-ai-videos-for-beginners-insane-results-from-3-prompts/27212/1 "2026-10-31T03:46:47Z")
[![](https://www.artofsm.art/uploads/default/original/3X/8/e/8ee43429a2404e015f8fac502df80d78298a255d.jpeg "How to Make AI Videos for Beginners: INSANE Results From 3 Prompts") ](https://www.youtube.com/watch?v=P6UNRh2pZBs) This tutorial guides beginners through creating impressive AI-generated videos using just three prompts on the Hicksfield platform, demonstrating how to generate detailed character images, multi-angle character sheets, and realistic animated videos with customizable settings. It also highlights features like Genjutsu for inserting custom characters into popular templates, emphasizing the ease and flexibility of AI video creation with practice and the right tools. ---
**Author:** ![merefield](https://www.artofsm.art/user_avatar/www.artofsm.art/merefield/32/74_2.png) [@merefield](https://www.artofsm.art/u/merefield)\ **Post date:** [31 October 2026 04:11 UTC](https://www.artofsm.art/t/how-to-make-ai-videos-for-beginners-insane-results-from-3-prompts/27212/3 "2026-10-31T04:11:29Z")
This tutorial demonstrates how beginners can create impressive AI-generated videos using just three prompts and no prior experience. The presenter introduces Hicksfield, a platform that hosts various AI models for image and video creation under one subscription plan. They explain the difference between image models, which generate still pictures, and video models, which animate those images. For this tutorial, the presenter uses GPT Image 2.5 for creating detailed character images and CDN 2.5 for producing realistic video animations, both accessible through the Hicksfield Plus plan. The first step involves creating a character image using GPT Image 2.5 with a detailed prompt describing the character’s appearance, clothing, background, lighting, and style. The presenter emphasizes the importance of crafting precise prompts and suggests using AI tools like Claude or ChatGPT to help draft them. After generating the initial character image, the next step is to create a character sheet showing the character from multiple angles—front, side, and close-up—to ensure consistency throughout the video. This is done by attaching the original image as a reference and generating a multi-panel image using the same model. Next, the tutorial moves to video creation using the CDN 2.5 model. The character sheet is uploaded as a reference, and a video prompt is written describing the character’s actions, camera movements, and audio. The presenter sets the video duration, aspect ratio, and quality before generating the video, which takes a few minutes. The resulting 15-second clip features the character performing actions and speaking lines, with attention to maintaining consistent appearance, outfit, and natural motion throughout the video. The presenter then demonstrates how to tweak the video prompt to change the setting and time of day, creating a new version of the video with the same character in a different environment. This highlights the flexibility of AI video generation, allowing users to experiment with various scenarios while keeping the core character intact. Users can download their favorite versions or continue refining prompts to achieve the desired result, making the creative process iterative and user-friendly. As a bonus, the tutorial introduces Genjutsu, a feature that lets users insert their custom characters into trending community video templates. This shortcut preserves existing motion and camera work while personalizing the content with the user’s character. The presenter encourages viewers to sign up through their link to receive discounts and access to unlimited trials of certain models, emphasizing that anyone can quickly learn to create AI videos with practice and the right tools. ## Useful Links - [Higgsfield AI Platform](https://higgsfield.ai/s/make-your-first-ai-video-yt-promptengineer48-hlAopr) — Central platform used for AI video creation in the tutorial. - [Seedance 2.5 Video Model on Higgsfield](https://higgsfield.ai/seedance/2.5) — Primary video model used to generate AI videos in the tutorial.