With A Little Help From My (Artificial) Friend

Original unedited scan from 35mm Kodachrome slide. (click to enlarge)

This photo above is one I made in 1996 at Monument Valley.  Just about everyone should recognize this scene.  For those of you who have never been there, I made this image while standing right next to the parking lot.  Things have since changed and when I returned last May, there is now a fence that has been erected and I don’t think this particular composition is possible.  In fact, this particular rock in the foreground may not even be there as well.  If it is, I wasn’t able to locate it as there is now a hotel, patio, railings, restaurant and gift shop occupying the area.

This photo shows an iconic scene from Monument Valley, certainly nothing a hundred thousand others haven't made.  It is, in and of itself, nothing special.  That said, I like it as it reminds me of my time there.  About 8 or 9 years ago, I had the 35mm Kodachrome slide commercially scanned.  Looking at this digital version you can see why I was disappointed by the result.  Although the original photo was sharp, the scan is not only not sharp but poorly scanned in every observable way.  Does that mean my photo lost forever?  Well, no.  I still have the slide and this particular scan is not good but another scan or being scanned in a different manner, such as the slide photographed with my 45mp Nikon Z8 camera, may produce a much better result.  But now there also is one other and much easier way that has emerged to rescue this image.

I wondered what one of the artificial intelligence (AI) image generating software programs  could do with it since every time I tried to edit this scan, I remained disappointed with the final result.  So, the other day I uploaded a copy of this image file, unedited and as you see above, to Google’s Nano Banana 2 AI Generative software—not to an image editing program.  Upon first response, it seems Nano Banana cannot edit the file directly so I asked it to create an identical file based upon the one I uploaded.  I wrote the following:

“Create a copy of the included uploaded image.  This photograph needs color corrected to faithfully reproduce the red rocks. The sky needs some scattered cumulus clouds. The shadows need added detail. The contrast and saturation need corrected. Finally, it needs sharpened so everything in the photograph is rendered sharp and detailed. In other words, make this photograph perfect.” 

At that point I hit ‘enter’ and it came back with a question for me.  It wanted to know if I also wanted it set in ‘the golden hour.”  Why not, I thought?  I answered, yes.

The image below is what it produced for me.  Amazing!  I think, with the addition of the golden hour lighting, long dark shadows and inclusion of some clouds to make the sky more interesting, this version is much better than mine.  And…I don’t think it could have been easier.  Also, it took about one minute to generate this new, completely artificial version.  The only thing I did was to upsize it in Topaz Gigapixel AI as the version Nano Banana produced was smaller than I typically post to this blog.

Monument Valley during Golden Hour.  Created entirely by Nano Banana 2. (click to enlarge)

The photographic lesson to be learned is, if you choose to do so, AI generative software can literally save one of your photographs if it is out of focus, off color, lacks visual impact, has blown highlights, lost shadows, etc.  In fact, you can take an ordinary photo at noon on a sunny day and have it transformed into just about any mood or look you desire!  No need to get up before dawn to drive out to your subject any longer.  

For my ‘real’ work, I won’t do this.  For experimentation and ‘playing around,’ I will utilize it to increase my knowledge base—to see what it can do.  Using AI like this is not why I photograph.  I photograph to exercise my creativity and for self-satisfaction.  In fact, my greatest satisfaction comes from the entire photographic process—finding an emotion evoking or interesting composition, determining and making the exposure according to how I want to capture what is before me, editing the image file to achieve how I envisioned the scene, printing it, mounting it, then framing and hanging the photograph on my wall.  For me, it doesn’t get any better than that.  

In an earlier post, I wrote about just this—utilizing artificial intelligence tools to make our photographs perfect and, even though tempting and enticing, it can have very negative consequences for the future of photography.  You can read that post here.

How about you?  Will you use AI to create or highly modify, add or change your composition or do you prefer to use it minimally?  This kind of capability is certainly going to greatly impact photography in the future.  

Join me over at my website, https://www.dennismook.com.  

Thanks for looking. Enjoy!  

Dennis A. Mook  

All content on this blog is © 2013-2026 Dennis A. Mook. All Rights Reserved. Feel free to point to this blog from your website with full attribution. Permission may be granted for commercial use. Please contact Mr. Mook to discuss permission to reproduce the blog posts and/or images.