Beyond the Frame

Exploring what lies beyond the borders of a photo with AI

Hi friend,

How are you? I hope you’re doing well.

Looking back, it may sound obvious now, but the art of photography has always been anchored in a single fact: capturing a fleeting moment in time and then allowing audiences to extend it further in their minds based on their narratives.

Nowadays, as a storyteller, with AI, you can always extend your original photo with your wildest ideas.

This way, audiences can see the full range of your imagination and creative associations. No worries, they’ll always add their interpretations.

In this sense, think of AI as a bridge between reality to fantasy.

From implicit to explicit articulation.

One thing you should know.

Because AI uses a generative model, any original photo you upload and ask it to augment will not add new elements to the exact original image, but rather statistically recreate your uploaded photo plus your new request.

At first glance, it may look the same, but expect variance.

Let’s look at some fresh examples:

Example 1

After a few rainy days in Miami the past week, I was finally able to take my morning walk around the neighborhood and got inspired by this view:

I liked how the red canopy of the Royal Poinciana is contrasted with the greenery all around and the red ground foliage. The yellow lines felt as if inviting us to discover a hidden realm. With this thought, I created a text prompt that transformed the red foliage into a longer red path leading to a magical vortex hidden within this quiet neighborhood.

Five years ago, animators would have spent hours creating each frame separately to materialize this visual concept.

These days, AI can help you not only polish your base idea into a stronger prompt but also visualize it in seconds.

For this very short clip, I provided a descriptive prompt and used the photo-to-video feature on Runway ML.

Example 2

Last weekend, we had a nice walk at South Pointe Beach in Miami Beach.

I came across this sign and, looking at all those prohibited Es, I figured that soon enough another E will join them:

So, with the help of ChatGPT 04-mini-high, I added the missing E and characters:

Notice the different icons and missing tiny bolts you see in the original photo? Also, the background is different. It’s not a replica but a very close approximation. For now…

Here is an earlier version of this photo, I turned into a video on Runway ML:

What do you think?

I especially like the nice touch at the end when a boy walks by with a pitying glance at the poor robots.

Ha! That was not even in the prompt…

Jokes aside, who knows, with the upcoming AGI, the roles could even be reversed :)

In short

This practice of enriching and extending base source content is an effective way to capture attention and bring your brand narrative to life.

I extended the first photo with a video and the second with an augmented photo, and then generated a video.

I know these are elementary examples, but with tools like Google’s new VEO 3 model, you can create much elaborate videos with dialogs, music, and special effects.

Here is a fun viral clip with a unique premise; characters are in disbelief they’re all made out of prompts…

Makes you think what Netflix would look like in 2030?

I can see a rich library of 100% AI-generated movies starring licensed copies of ageless Tom Hanks and Julia Roberts, flanked by a classics section - all done the old way.

In fact, Robert Zemeckis’ recent movie Here has already applied AI featuring younger and older Hanks.

One more thing, when generating your AI content, the essential ingredient you need to bring with you beyond your idea is loads of patience.

Chances are, your first “pancake” won’t be perfect. Heck! Even the fifth…

It’s a generative model, remember?

Just to get the icons on the sign to stay closer to the original photo, and keep the E-ROBOTS icon meaningful, was a struggle (see in the above photo and video).

I left it as is - call it my imperfection thread.

It’s a very iterative process because, let’s face it, you’re in essence sitting in the passenger seat just giving instructions to a very opinionated AI driver…

See you next time!

Best,

- Shlomi

Shlomi Ron
Founder, Visual Storytelling Institute
shlomi@visualstorytell.com
story > visual > emotion > experience

Loved this post? Take your visual storytelling skills to the next level as a Narrative Partner. You’ll get:

• The “Top 10 Business Storytelling Formulas” eBook—practical frameworks to craft marketing that captivates and converts.

• Exclusive content + bi-weekly Sunday insights to sharpen your narrative thinking, decode trends, and apply visual storytelling with confidence.

• Full archive access for endless inspiration, fresh ideas, and proven techniques—ready whenever creativity strikes.

• A quarterly “30-minute” personalized coaching session to up your visual storytelling game.

Start your 7-day free trial and grow into the storyteller your brand needs.