"black background, 4k resolution, extreme detail, ultra high resolution: real life living anatomically accurate butterfly in flight covered in precious gemstones and crystals. diamonds, emeralds, rubies, sapphire, etc. 4 point diffraction spikes, subsurface scattering, chromatic abberation, dispersion rainbow. macro shot, professional photography, dramatic lighting, directional lighting from left, hard light, hard shadow, rim lighting. butterfly is standing on a finger"
Google Gemini (Nano Banana) > Google VEO 3.1
Looking to the Future
THE POWER
OF THE PROMPT
I have been what is now called "prompt engineering" since AI image generation was considered no more than a fidget toy, with no real world application. Starting with MidJourney in early 2022, through the introduction of ChatGPT 4o in 2025, and now with Google's Nano Banana & VEO 3.1, I have been closely watching all the latest AI technologies and refining my prompts for the past 4 years.
Because of my multidisciplinary approach to design, I have been able to incorporate all of these disciplines into creating prompts, pulling from 3D material design, basic principles of photography, illustration techniques, and more general design principles. This has allowed me to communicate exactly what I want to the model in immense detail, in turn getting 4k, high production value assets at scale, fully in line with brand guidelines.
%20(1).png)
MULTI-CHANNEL
APPLICATION
Since the technology became scalable in 2025, my AI generated imagery has been used across multiple channels, including web imagery, retention marketing, and acquisition, specifically in static and video ads.
When designing static ads, I have reached the point where I am able to generate a new, high production value asset for each individual static ad, with minimal impact on production time.
REUSE, RECYCLE
Besides generating unique images, I am also able to use this technology to create variations in existing images. I have found this to be extremely valuable in creating site product imagery, which utilize many of the same, repetitive elements. It is also extremely valuable in performance creatives, as I can turn one image from a winning creative into many, each with subtle changes, ready to be used for testing variations.
ETHICS
Given the nature of this technology, and its ability to spread misinformation, I have not and will never generate imagery that is not a 100% faithful representation of the original product. This is in part why I waited until the models matured enough in capability to begin using them at scale.
I also believe this technology should be used to supplement and expand on photography and videography, not to replace it. I see it more as an advanced photo manipulation tool, and use it as such.
