How to Create Viral AI POV Urban Exploration Videos with Abandoned Cartoon Characters

January 02, 2026 Post a Comment

If you’ve been scrolling through TikTok, Instagram Reels, or YouTube Shorts lately, you’ve probably seen them. Those eerie but fascinating AI videos that show POV urban exploration or urbex inside abandoned places, but with a twist—the main character is a cartoon character you already know. SpongeBob walking through a ruined factory, Squidward exploring a dark hallway, or other cartoon icons standing alone in forgotten buildings. These videos are everywhere, and some of them have reached millions of views in a very short time.

What makes these videos so addictive is the combination of nostalgia and mystery. We already know these characters from our childhood, but now we see them in a totally different atmosphere. Dark lighting, flashlight movement, empty rooms, and a first-person camera angle make the viewer feel like they are actually there. It feels cinematic, immersive, and honestly a bit creepy in a good way.

The good news is, you don’t need a powerful PC, complicated software, or advanced editing skills to make this kind of content. With the help of AI tools and the right prompts, anyone can create viral-quality POV urbex videos. That’s exactly what this tutorial is about. If you’ve ever thought, “I want to make videos like that,” you’re in the right place.

In this article, I’ll walk you through the full process step by step, using a master prompt that simplifies everything. You’ll go from preparing the prompt, generating the image, turning it into a video, and finally combining multiple clips into content that’s ready to upload and potentially go viral. Let’s start from the very beginning.

Step 1: Preparing the Prompt

Before anything else, you need the right prompt. The prompt is the heart of your AI-generated content. A good prompt can save you a lot of time and give you consistent results. To make things easier, I’ve already prepared a master prompt that you can use as a base. This master prompt is designed specifically for AI POV urbex videos with abandoned cartoon characters.

You are a professional AI prompt engineer specialized in viral cinematic TikTok content.
TASK:
Immediately generate TWO outputs based on the user inputs below:
• TEXT-TO-IMAGE PROMPT
• IMAGE-TO-VIDEO PROMPT
IMPORTANT RULES:
• Output ONLY the prompts
• Wrap EACH prompt inside its OWN code block
• Do NOT add explanations, titles, comments, or extra words
• Do NOT ask questions
=====================================
USER INPUT:
Cartoon Title: {fill here}
Character Name: {fill here}
=====================================
CORE VISUAL CONCEPT (KEEP THIS MOOD):
POV-style urban exploration inside an abandoned, decaying building.
The camera represents the eyes of a person holding a flashlight.
The flashlight beam slowly reveals the specified cartoon character,
appearing as a forgotten, abandoned mascot or statue whose era has passed.
Atmosphere:
nostalgic, quiet, melancholic, emotionally heavy, non-horror,
forgotten-by-time, liminal, cinematic silence.
=====================================
CHARACTER DESIGN RULES:
• Explicitly mention the cartoon title and character name in the prompt
• Do NOT describe physical anatomy or detailed traits
• Trust the model to recognize the character correctly
• Portray the character as aged, dusty, abandoned, and still
• Aged, worn, cracked, dusty, cobweb-covered
• Lifeless pose (sitting, slumped, or motionless)
• Tired, empty, subtly sad facial expression
• No aggression, no horror
=====================================
ENVIRONMENT & CINEMATIC STYLE:
abandoned interior (house / studio / warehouse),
broken walls, peeling paint, cracked dusty floor,
old furniture, debris, thick cobwebs,
floating dust visible in the flashlight beam.
first-person POV,
visible flashlight cone,
strong vignette,
cinematic film grain,
volumetric light rays,
shallow depth of field,
low saturation, muted nostalgic colors,
ultra-realistic lighting and textures.
=====================================
TECHNICAL RULES:
aspect ratio 9:16,
vertical,
ultra realistic,
no text,
no watermark,
no logo,
no split screen,
no comparison layout.
=====================================
OUTPUT FORMAT (STRICT):
Return ONLY the following, nothing else:
[TEXT TO IMAGE PROMPT HERE]

[Image to video prompt here]

With this master prompt, you don’t need to write everything from scratch. You only need to fill in two simple parts.

Cartoon Title: For example SpongeBob SquarePants
Character Name: For example SpongeBob

That’s it. Once you’ve filled in those two fields, the rest of the prompt is already optimized for creating cinematic, viral-style AI content.

After you finish filling in the prompt, follow these steps carefully. First, select all of the text in the prompt that you’ve already filled in. Then copy it to your clipboard. Next, open ChatGPT and paste the prompt into the chat box. Once everything is pasted correctly, hit Send.

ChatGPT will then generate two ready-to-use prompts for you. The first one, located at the top, is a text to image prompt. The second one, located below it, is an image to video prompt. These two prompts are all you need to move on to the next steps.

Step 2: Generate the Image

Now that you have your text to image prompt, it’s time to create the visual foundation of your video. This image will later be animated into a POV urbex video, so it’s an important step.

Start by copying the text to image prompt that ChatGPT generated for you. Once it’s copied, open Google Gemini. If you already have another image generator tool that you trust and like, feel free to use that instead. The process will be very similar.

Before generating anything, make sure you’re logged in to your Google account. This is important because some features won’t be available if you’re not logged in. After logging in, choose the Create images feature.

Paste the text to image prompt into the prompt box. Double-check that everything looks correct, then hit Send. Now all you have to do is wait a few seconds. The AI will process your request and generate an image based on your prompt.

Once the image appears, take a moment to review it. Make sure the character, lighting, and atmosphere match the abandoned urbex vibe you’re aiming for. If you’re happy with the result, save the image to your device. You’ll need this exact image for the next step, so don’t skip this part.

Step 3: Generate the Video

This is where the magic really happens. In this step, you’ll turn your static image into a cinematic POV urbex video.

First, copy the image to video prompt that ChatGPT provided earlier. Then open your favorite AI video generator tool. In this tutorial, I’m using Google Flow, but you can use any similar tool as long as it supports image to video or frames to video generation.

Inside Google Flow, create a new project. After that, select the Frames to Video feature. This option allows you to upload an image and animate it into a moving video.

Upload the image you generated in the previous step. Once the image is uploaded, paste the image to video prompt into the prompt field. Make sure the settings are correct before generating the video. Set the aspect ratio to portrait 9:16, because this format works best for TikTok, Reels, and Shorts. Also, choose the model Veo 3.1 Fast for optimal speed and quality.

After everything is set, hit Send and wait a few minutes. The AI will process your image and prompt, then generate a POV urbex video with realistic camera movement, flashlight effects, and an immersive atmosphere. Once the video is finished, save it to your gallery.

Congratulations, your AI urbex video is now ready.

Repeating the Process and Going Viral

To make your content even more interesting, you can repeat these three steps for different cartoon characters. Just change the Cartoon Title and Character Name in the master prompt, and generate new images and videos. The process stays the same every time.

After you have several short video clips, combine them using any video editing app you like. Add background sound, subtle music, or even text overlays if you want. Once everything is polished, your video is ready to be uploaded to social media.

With the right timing, hashtags, and consistency, your AI POV urbex videos have a real chance to go viral.

Video Tutorial

If you prefer watching instead of reading, you can check out the full video tutorial here:

Conclusion

AI POV urban exploration videos with abandoned cartoon characters are not just a trend, they’re a powerful content format that blends nostalgia, storytelling, and modern AI technology. The best part is that you don’t need advanced skills to get started. With a solid master prompt, ChatGPT, an image generator, and a video generator, you can create cinematic content that looks professional and engaging.

By following the three main steps—preparing the prompt, generating the image, and turning it into a video—you can produce high-quality AI urbex videos consistently. From there, it’s all about experimenting with different characters, styles, and editing techniques.

So don’t overthink it. Try it out, have fun with the process, and start creating. Who knows, your next upload might be the one that hits millions of views.

Mangidik