Before Mount Rushmore: The Story of Spotted Tail

The Spark

A friend of mine works with the Rosebud Sioux Tribe in South Dakota. After seeing some of my AI work, he posed an intriguing challenge: "Why don't you create a concept and pitch it to tribes to help with historical preservation?" The idea was to offer something more cost-effective than a Hollywood production, yet still powerful and respectful, something that could preserve culture and inspire a new generation to see their history in a way they'd never experienced before.

The initial plan was simple: create a 15-second proof of concept. But I made one critical decision that changed everything... I started researching.

All I had was a name: the Rosebud Sioux Tribe. Within hours of digging into their history, I discovered their most important historical figure: Chief Spotted Tail (Sinte Gleska), a Sicangu Lakota leader whose story deserved so much more than 15 seconds. I was hooked. What began as a simple concept became a complete immersion into his legacy, and I dove headfirst into what would become a 110-hour journey.

The Research Phase

I spent approximately 30 hours in deep research, watching documentaries, reading accounts from the period, studying not just Spotted Tail, but the broader context of Lakota culture, Native American history, and even other tribes to truly understand the heritage I was honoring. I knew that if I was going to tell this story, I had to do it right.

During this phase, I bounced between multiple AI platforms, mostly ChatGPT and Gemini, testing ideas, refining concepts, and ensuring every element felt authentic and strong. Some ideas didn't land immediately, but I kept them in my notes, knowing I'd return to strengthen them later.

From this research emerged the core elements: the narrative structure, the poetic narration style, and the emotional arc I wanted viewers to experience.

Building the Foundation

With my research complete, I began the creative build:

Voice Creation: I spent about three hours in ElevenLabs crafting the perfect voice. One that could carry the weight and reverence this story deserved.

Music Selection: I discovered traditional Lakota music and found recordings from a 1970s Pow Wow. The Veterans/Victory song, which predates First Contact, became the backbone of the main piece. Later, I would add the Sioux flag song, a more recent composition that serves as a national anthem for the Sioux people, to accompany the historical credits section.

Musical Adaptation: Using SUNO, I created covers of these traditional songs as references, working to produce cinematic adaptations that would be both powerful and respectful of their cultural significance. Together, the songs frame the video as a tribute: the Victory Song energizes the core story of individual heroism, while the Flag Song solemnizes the collective historical context, reinforcing Lakota identity and respect for leaders like Spotted Tail.

The Production Process

Once I had my framework, the real work began. I mapped out the video shot-by-shot, detailing every beat, every visual, and the emotional tone each moment should carry. I work best when I can see the full shot list on the timeline as subtitles, starting with the music and audio foundation, then building the visuals systematically, one shot at a time.

This was harder than I anticipated. I wanted the piece to flow seamlessly. I wanted cohesion. I wanted every idea to work in harmony with the others. While trying to capture key elements of a complex 60 years history. Achieving that required far more hours than I'd expected.

As I neared completion of the main video section, I had a realization: the poetic approach, while beautiful, wasn't enough. People might feel the emotion, might connect with the imagery, but they wouldn't truly understand what happened to Spotted Tail and his people.

The History

So I decided to add an entire straightforward historical breakdown that would give viewers the core, unvarnished history of Spotted Tail and the Lakota people, along with my deep appreciation for their legacy. While also sharing real historical photos and some elements of AI restoration/recreation as well.

For this credit-sequence-style section, I shared real historical photos together with created AI-generated recreations of of those photos as well as adapting some of the figures to cinematic video representations of those people, visual anchors that could draw viewers in even if they didn't read every word of the written history section. I wanted multiple entry points for different types of viewers: those who would read the history, and those who would connect through the visuals alone.

Seeking Excellence

By this point, I had invested approximately 80-90 hours into the project and had already done two significant overhauls, each time rebuilding major sections when something didn't feel right. But I wasn't satisfied with "good enough." I wanted to do justice to this story and push myself to deliver the best work I was capable of at that time.

I reached out to a friend who runs a small studio, asking for honest feedback on what worked and what could be stronger. His insights were invaluable, and I incorporated several of his suggestions.

But I wanted to go further. I knew of Kavan the Kid, recognized as one of the top AI film directors in the space. I decided to invest in myself and hired him for an hour-long consultation.

A few days before that call, I nearly scrapped the entire project.

I was at about 97-98% completion, but all I could see were flaws. "I can only see the bad in this," I told Kavan. "I see 50-60 things that bother me, but if I took the significant time to fix those then I would just find more things that aren't perfect and then those would bother me too."

His response calmed me down: "That is the creative process."

He explained that this feeling, this hyper-focus on imperfections, happens on all his projects too. The difference is having a deadline and committing to release no matter what when you reach it. That perspective resonated deeply with me.

Kavan's technical feedback was precise and transformative. He confirmed I was already there—the piece was very strong, and he identified the specific polishing touches that would elevate it from strong to top-tier. I implemented every suggestion.

The Finish Line

110 hours. That's what it took.

At a certain point, I had to accept that I could continue refining indefinitely. Perfection is a moving target. So I made the hardest decision of any creative process: I put the bow on it, and released it into the world.

The Reception That Mattered Most

The feedback from the AI community was encouraging, respected voices in the space recognized the quality and effort that went into the piece.

But the reception that truly mattered came from the Rosebud Sioux Tribe themselves.

Professor Royal Lost His Blanket-Stone, head of the Lakota Studies Department at Sinte Gleska University (a university named after Spotted Tail himself), reached out personally. He called the piece "very impressive," appreciated the historical accuracy and respect, and noted that it could be valuable to their department going forward.

Victor Douville, another member of the Lakota Studies Department, was deeply moved by the video. He asked if I would share the AI-generated images of Spotted Tail so he could use them in his own work. This feedback, from the people whose culture and history I had tried to honor, validated every hour, every iteration, every moment of doubt.

The Challenges

The challenges weren't just technical, they were multi-layered and interconnected.

The Moving Target of Technology: One of the most unique challenges of working with AI video is that the technology evolves while you're creating. During this project, new tools and features kept releasing: Kling 2.1 First Frame/Last Frame, Nano Banana, and others. Each release tempted me to remake everything, to leverage the latest capabilities. I had to learn to take a deep breath, step back, and make strategic decisions about what to adopt and what to leave alone. I used some new tools for later sections, but I resisted the urge to rebuild large portions of completed work. Otherwise, this project would never have ended.

The Scope Creep Monster: What started as a 15-second proof of concept progressively grew and grew and grew. Each layer of research revealed new aspects of the story that deserved telling. Each creative decision opened up new possibilities. The project expanded organically, but without clear boundaries, I found myself feeling pressure, stuck, and boxed in by plans I had made when the project was originally smaller.

The Technical vs. Conceptual Balance: Looking back, I realize I did solid work on the conceptual planning: the narrative arc, the emotional beats, the cultural respect. But the technical planning needed more attention upfront. Too often, I wanted to add more information or expand the story, only to realize the music timing or narrator pacing wouldn't accommodate it. I felt constrained by earlier decisions that were now difficult to change without cascading effects throughout the entire piece.

The Creator's Paradox: There was a challenge to persist. There was a challenge to continue improving even when the piece was already good. There was a challenge to not give up when all I could see were the 50-60 imperfections. To bear down and give everything when 60 or 70 or 80 hours could have been enough.

But the biggest challenge was knowing when to stop, when to trust that I had given this story my heart and soul, even when my perfectionist brain screamed that it wasn't ready.

What I Learned

This project taught me that meaningful work requires more than skill, it requires surrender to the process. It requires honoring the subject matter above your own convenience. It requires seeking guidance from those better than you and having the humility to implement it.

I learned that 110 hours isn't too much when the story matters.

I learned that cultural respect isn't just about research, it's about responsibility, about understanding that you're temporarily holding someone else's legacy in your hands.

But I also learned practical lessons that will shape every future project:

Define the scope early, or it will define itself. What began as 15 seconds became a multi-section film because I let the project grow organically without setting boundaries. This felt very restricting when it came to wanting to change or add information. I was happy about the work I did on the narrative arc, the emotional resonance, the cultural authenticity. But I underinvested in technical planning. When I wanted to expand the story later, I found myself constrained by music timing, narration pacing, and structural decisions that were now costly to change. For future projects, I'll map out the technical framework with the same rigor I apply to creative development and try to find a way to build flexibility into the process by maybe rearranging some of the steps in my workflow. This will take some testing to nail down properly.

The creative process might mean you'll only see the bad. Kavan's words about this being a common hurdle that many of the top creatives face allowed me to give myself permission to release work even when I could see 50-60 imperfections. Perfection isn't the goal; completion is. A positive element to this, at times, torturous process is that it draws out the best of you. Forces you to dig deep and push hard and continue to improve when others would settle.

But having a deadline and honoring it, no matter how much you want to keep polishing, is a discipline I now understand is also essential.

New tools will always emerge; chasing them will trap you. In AI video, technology progresses faster than most project timelines. Learning to strategically adopt new capabilities without falling into the remake trap is a skill in itself. Some updates are worth integrating, others are distractions that will prevent you from ever finishing.

And I learned that the most valuable feedback doesn't come from awards or recognition, it comes from the people whose stories you're trying to tell, saying: "Thank you. You honored us well."

This project became a benchmark for every AI project I'll create going forward. It showed me what's possible when you refuse to compromise, when you let yourself be transformed by the research, and when you give everything you have to give.

Before Mount Rushmore: The Story of Spotted Tail isn't just an AI video to me.

To me, it was testament to how this technology doesn't only propel us forward, it gives us unprecedented power to look backward, to illuminate our past with new clarity, and to forge emotional connections with history in ways that transcend what books alone could achieve. It allows us to reach the hearts of youth and modern audiences who might never open a history text, meeting them where they are and showing them their heritage in a format that resonates with how they experience the world.

Tools:

Research/Script:

• ChatGPT o3 and 5 Thinking
• Gemini 2.5 Pro Deep Research and Deep Thinking
• Grok 4 Expert
• Perplexity and Comet

Prompts:

• ChatGPT o3 and 5 Thinking (Custom GPT)
• Grok 4 Expert

Image Gen/Editing :

• Google Imagen 4 Ultra and Nano Banana
• Ideogram v3 Character Reference/Magic Fill
• Midjourney
• ChatGPT SORA (image)
• Black Forest Labs Flux Kontext on ComfyUI
• Leonardo Ai Lucid Origin

Video Gen/Animation:

• Google Veo 3 Fast and Quality
• Hailuo 02, Kling 2.1
• Bytedance Seedance 1.0
• Runway Aleph

Upscaling:

• Magnific Precision
• Topaz Labs Video AI, Astra, Bloom

Audio:

• SUNO
• Eleven Labs SFX and Voice Creator
• Artlist IO SFX
• Adobe audio enhance v2 (only used on like 2 words)

Misc/utility:

• Adobe Firefly boards (for Nano Banana)
• Holy Grain Film Grain Overlay

Editing:

• Capcut
• DaVinci Resolve

Built With

• adobefireflyboards
• artlistio
• astra
• capcut
• chatgpt5thinking
• chatgpto3
• chatgptsora(image)
• davinciresolve
• elevenlabs
• fluxkontext
• gemini2.5pro(deepresearch/deepthinking)
• googleimagen4ultra
• googlenanobanana
• googleveo3(fast/quality)
• grok4expert
• hailuo02
• ideogramv3
• kling-ai-2.1-bytedance-seedance-1.0-on-wavespeed-@runwayml-aleph-upscaling:-magnific-precision-@topazlabs-video-ai
• kling2.1
• leonardolucidorigin
• magnificprecision
• midjourney
• perplexity
• perplexitycomet
• runwayaleph
• seedance1.0pro
• suno
• topazlabsastra
• topazlabsbloom
• topazlabsvideoai