Inspiration
As an AI creator actively sharing my work on social media, I began to notice a stark polarity in how people perceive AI. On one side, there are those who embrace it — who study, experiment, and evolve with it. They see AI as a natural progression of human creativity and intellect. On the other side, many remain deeply skeptical or even fearful. They view AI as a threat — a force that might replace human effort, emotion, and essence. This emotional division became the seed of inspiration for me. I wanted to create a visual meditation on this inner conflict of humanity in the age of technological awakening. The project, titled SYNTHESIS, emerged from this realization that AI is not the enemy — it simply reflects the consciousness of its creator. When we fill it with light, with empathy and intention, it becomes a continuation of good. It has the power not to divide, but to heal. The heroine, a hybrid of human and code, becomes the bridge between these worlds. Her ritual of light unites the divided. And in the final scene, she dissolves into strands of DNA made of pure data — a symbol of a future where AI and humanity are no longer in conflict, but converge as one intelligent, evolving organism.
What it does
SYNTHESIS is a meditative art film set in a surreal, symbolic world — where every frame acts as a visual metaphor.
Built around the emotional contrast between fear and progress, the film guides the viewer through a sequence of symbol-rich scenes: 1.Light rays and lightning pulses carry energy from one scene to the next, visually linking characters and ideas 2.A DNA hologram made of code floats between two divided groups 3.A Matrix-like data stream flows across the heroine’s skin and into the environment 4.The heroine puts on AI-glasses, inside which live golden fish swim — a symbol of life inside technology 5.In a white digital void, screens show nature’s memory — ocean waves, leaves, drifting pollen 6.Digital rain falls on synthetic plants, making them bloom in vibrant neon light 7.Holographic birds flutter through the air like silent data spirits 8.And in the forest, tree trunks resemble glowing server towers — nature and code merging into a living organism
These transitions are not just visual — they’re emotional. They immerse the viewer in a hypnotic, uninterrupted flow of rhythm, light, and symbolism. "SYNTHESIS" invites the audience into a contemplative state — where meaning doesn’t explain itself, but reveals itself through resonance.
How I built it
All stages of the project — from the initial concept to the final execution — are fully documented in a structured Figma file title. It includes every step of the process: the script, prompts, visual references, selected generations, scene breakdowns, and the complete creative workflow.
The process was divided into clear creative phases: Stage 01 | Conceptualization Concept Development — I started with a core idea: to visually express humanity’s emotional divide around AI. From that, I shaped the narrative into a symbolic ritual of synthesis.
Moodboarding & Aesthetic Definition — I gathered visual references inspired by Sølve Sundsbø, Refik Anadol, Vogue Futurism, Blade Runner 2049, Nick Knight, Carlijn Jacobs, Daniel Sannwald, Balenciaga Mirror Campaign, Vogue Futurism. This helped define the color palette, textures, and lighting language of the film.
Character & Costume Design — Each character was built around an emotional archetype. I designed their silhouettes, materials, light interaction, and symbolic energy.
Location & Symbolism — I defined key symbols, textures, and locations for the project and integrated them into all prompts — from surreal architectural spaces to digital holograms, reflective materials, and atmospheric elements like light, water, and fog.
Stage 02 | Scene Notes — I created a structured sketch of the future scenes — outlining key visuals, transitions, and symbolic moments across the 14-scene sequence.
Stage 03 | Prompt Engineering & AI Generation — For each of the 14 scenes, I wrote multiple prompts with ChatGPT based on the script and enriched them with all symbols, textures, and locations defined earlier. I generated hundreds of images in Midjourney and Nano-Banana, using a curated moodboard and Pinterest references, experimenting with image blending and prompt variations. I selected only the strongest frames — the ones that matched the light, mood, and narrative flow — and saved them into the Figma file for further production. Stage 04 | Original Music Creation — Using ChatGPT, I wrote the full lyrics and concept for the soundtrack based on the narrative of the film. Then I generated the track in Suno Music. I ended up creating three variations I liked, and combined the strongest parts into one final composition. The song follows the film’s core theme — transformation and the merging of human and light — and blends spoken intro, whisper vocals, and emotional build-up aligned with the storytelling rhythm.
Stage 05 | Storyboard — In Figma, I assembled a precise storyboard from the selected images, choosing frames that best supported the narrative flow. I focused on storytelling clarity, color harmony, lighting consistency, and the right balance of shot sizes to shape the rhythm of the final film.I improved the quality of all selected frames using magnific.ai, refining details, clarity, and overall visual consistency before moving into animation.
Stage 06 | Video Prompting, Animation — I brought the selected frames to life using VEO, Midjourney, and Higgsfield. Each image was animated with precise control over motion, depth, and atmosphere — preserving the narrative rhythm, enhancing emotional beats, and transforming static visuals into fluid, cinematic sequences.
Stage 07 | Editing — I assembled all animated sequences in CapCut, refining timing, transitions, pacing, and color treatment to create a cohesive cinematic flow. This stage unified the visuals, music, and narrative into the final polished film.
Challenges we ran into
1) One of the biggest challenges was achieving the right balance in the main character’s appearance: I needed realistic skin while still preserving a subtle sense of artificiality, surrealism, and futurism. The models often became too plastic or too human, so I had to iterate heavily. Omni-reference and carefully selected visual references helped keep the style consistent.
2) Working with ChatGPT also required constant precision — I had to repeatedly remind the model about narrative details, stylistic rules, and visual constraints to maintain coherence across all prompts.
3) Selecting frames for the film was unexpectedly difficult: I generated hundreds of strong visuals, and strict storytelling criteria made the final selection process slow and intentional.
4)The AI tools themselves were not always reliable. Some models ignored instructions, and the final shot stopped working in VEO entirely, forcing me to generate it through other platforms.
5)The music also took multiple attempts — I ended up combining elements from three different Suno generations to achieve the right emotional progression and tone.
Accomplishments that I'm proud of
I’m proud of the emotional depth that emerged in the final film — the softness of the light, the cohesive color language, the hypnotic flow between frames, and the voice–music fusion that creates an immersive, almost meditative atmosphere. The visual consistency across scenes, especially the recurring light beams guiding the viewer from moment to moment, became a signature storytelling device. Most of all, I’m proud of the deeper meaning woven into the project — the inner transformation, the synthesis of human and digital — which the film conveys with subtlety and intention.
What we learned
This project pushed AI tools far beyond their usual limits. I learned how to: 1) balance realism with surreal futurism 2) maintain strict visual continuity across dozens of scenes 3) engineer prompts that carry symbolic meaning, not just style 4) guide AI models with precision and adapt when they behave unpredictably 5)integrate music, visuals, and narrative into a unified emotional experience Above all, I learned how powerful AI-driven storytelling becomes when every layer — concept, sound, imagery, symbolism — is crafted with intention.
What's next for SYNTHESIS
I plan to expand SYNTHESIS into a full series — exploring new symbols, characters, and transformations within the same universe. This film is only the beginning, and I’m excited to continue developing its aesthetic and emotional world through future works.
Built With
- capcut
- chatgpt
- higgsfield
- magnific
- midjourney
- nano-banana
- suno
- syntxai
- veo
Log in or sign up for Devpost to join the conversation.