Meet Bella
YOUR GUIDE TO
STOP. CHECK. TELL.
scams and protecting your account.
- Project
- Project Type
- Project Status
- My Role
- Target Audience
- Tools
- Focus
- Timeline
- Blog Post
- Stop. Check. Tell.
- Academic AI-Assisted Educational Video
- Completed Academic Project
- Concept Developer, Content Strategist, Visual Designer and Video Producer
- Roblox Players Ages 13–16
- Freepik AI Studio, Kling AI, ElevenLabs, Sync.so, Suno, Adobe Audition, Adobe After Effects and Adobe Media Encoder
- Media Literacy, Scam Awareness, Audience-Centered Messaging and AI-Assisted Video Production
- Approximately One and a Half Weeks
- Read the Full Project Walkthrough
Project Details
Stop. Check. Tell. was created as the final project for a college course focused on misinformation. The assignment challenged me to develop a media message that would help a specific audience recognize how misleading content tries to influence their decisions.
I created a short AI-assisted video for younger Roblox players about fraudulent offers promising free Robux. The message focuses on a simple response viewers can remember and use when an online offer feels suspicious: stop before clicking, check the source, and tell someone they trust.
I developed the concept, audience approach, educational message, visual direction, and final video production. Bella was created as a peer-style presenter who could communicate the lesson in a format that felt familiar and approachable to younger viewers.
Looking Ahead:
Created in 2025, this video reflects both the possibilities and limitations of the AI tools and production workflow I used at the time. Reviewing it now, I see opportunities to improve Bella’s lip-sync, facial movement, and overall delivery.
As AI video technology continues to improve, I plan to revisit this project as time allows and recreate it using newer tools. Comparing the original and updated versions will provide a practical look at how the technology has changed, how my process has developed, and which improvements make the greatest difference to the finished message.
Production Workflow
I combined generative AI platforms with Adobe production tools to create the character, voice, movement, music, and final composition. Each tool handled a specific part of the project while I directed the message, visual style, pacing, and final presentation.
1. Message and Script:
I developed the “Stop. Check. Tell.” framework and shaped the script around one clear lesson: pause before responding to suspicious offers, verify the source, and ask someone trustworthy for help.
2. Character Design:
I used Freepik AI Studio to create Bella as a friendly, peer-style presenter. Her appearance needed to feel appropriate for younger viewers and fit naturally alongside gaming-inspired visuals.
3. Voice, Movement and Lip-Sync:
I generated Bella’s voice with ElevenLabs, created her movement with Kling AI, and used Sync.so to coordinate the narration with her lip movements.
4. Audio and Scene Production:
I cleaned and edited the narration in Adobe Audition and generated the background music with Suno. I balanced the audio so the music supported the energy of the video without competing with the message.
5. Editing and Delivery:
I assembled the character, narration, music, text, graphics, and animated scenes in Adobe After Effects. I then used Adobe Media Encoder to export the completed video in a format suitable for online viewing.
Challenges and Creative Decisions
This was my first time creating an AI-assisted educational video with this combination of tools. The project required me to learn new platforms, develop an audience-focused message, and complete the production within approximately a week and a half.
Working Within a Short Timeline: I began the project with limited experience using this type of production workflow. I needed to learn each platform while continuing to develop the character, narration, animation, and final composition within the assignment deadline.
Coordinating Multiple Tools: The character, voice, animation, lip-sync, music, audio editing, and scene production came from separate platforms. I had to bring those outputs together while maintaining a consistent visual style, pace, and tone.
Designing for Younger Viewers: The subject involved a serious online risk, but the video still needed to feel approachable to teenagers. I chose a peer-style presenter, energetic pacing, and gaming-inspired visuals to make the message feel relevant without making the scam appear harmless.
Simplifying the Message: Rather than overwhelming viewers with a long list of warnings, I centered the video around three memorable actions: Stop. Check. Tell. This gave younger viewers a practical response they could recall when faced with a suspicious offer.
What I Took From It
This project reinforced that audience strategy should shape every production decision. The character, pacing, language, music, and visual style all needed to support the habits and expectations of younger viewers rather than simply reflect my own creative preferences.
I also gained practical experience coordinating several AI platforms with traditional Adobe production tools. While the generative tools helped create individual assets, the final message still depended on planning, editing, quality review, and a consistent creative direction.
The strongest lesson was that an educational message does not need to be complicated to be useful. By reducing the response to three clear actions, Stop. Check. Tell., I created a framework that viewers could understand quickly and apply beyond a single Robux scam.