Well, here we are. Another week down and another huge AI innovation pushing this budding technology to another new level. Quite similar in many ways to Adobe Firefly’s Generative Fill tool, Midjourney has released a new v5.2 update that includes a Zoom Out prompt which is leaving many creators—again—flabbergasted at what this AI technology can create.

Let’s take a look at this new Midjourney update and explore this new Zoom Out prompt in particular to understand a bit about how it works and what it could offer. Plus, for those looking ahead to the future of AI for video, let’s take a stab at guessing where new tech like this might eventually settle into video production and editing.


Zoom_out_promptA look at the Zoom Out prompt in action.Credit: Midjourney

Midjourney v5.2 Update

Currently the newest and most advanced version of Midjourney, a generative AI program and service created and hosted by the Midjourney, Inc. research lab in San Francisco. This latest version—like the rest of the AI generators on the markets—is still using prompts similar to DALL-E 2 or Stable Diffusion to generative images from natural language descriptions all through their popular Discord server.

Released here in June 2023, this new model is accessible on their servers to users by simply adding the “v5.2 parameter” to the end of their prompts (or by going into the server’s settings and selecting the latest version). 

Along with the Zoom Out prompt feature, which we’ll get into below, this v5.2 update should see a good deal of sharper detail results with better colors, contrast, and composition as Midjourney’s AI continues to learn and refine its image generation.

You can read more about the full v5.2 update here on Midjourney’s homepage.

Midjourney Zoom Out Prompt

While the stylistic upgrades to Midjourney are indeed pronounced, the biggest feature update is this new Zoom Out prompt. As its name implies, this Zoom Out tool will allow users the ability to prompt the Midjourney AI to “zoom out” on an image they have uploaded or created to provide further context generated by the AI.

Midjourney is adding this Zoom Out tool as a button underneath their generated images now and users are presented with two options for a “Zoom Out 1.5x” or “Zoom Out 2x” to decide how much more coverage they want generated. Midjourney has also included a “Make Square” option which uses the same prompt tech to turn any non-square image into a square one, which should be appealing to those looking to Instagram-ize their images.

You can see this Zoom Out prompt in action in some of these Twitter threads below.

Zoom Out for Video

However, while this is indeed impressive tech, as we mentioned this isn’t too different from Adobe Firefly’s Generative Fill tool or some others we’ve seen already in this space. Yet, with Midjourney being one of the most popular platforms for those experimenting with generative AI, several intriguing aspects to this new development may lead us to some theories about the upcoming AI for video.

As you can see in the example above, by adding so much more depth and context (which you can expound upon as you zoom out multiple times per image) you can create quite a large amount of information. Even with just a still image for now, with this much information and space to deal with, you can now start adding camera movement and zooms to any 2D image to give it more dynamism. 

This is a technique we’ve seen already in those AI-created fake film trailers, as a simple, yet effective, way to turn your prompts into something much more cinematic and video-like.

What’s most appealing—or concerning—though has to do with just how fast this technology came about. It’s one thing for an AI to generate an image based on a prompt. We’ve seen some video versions of this with RunwayML (which we’ve covered here), however with Midjourney so quickly adding more tools and features to their basic image gen prompt, it wouldn’t be surprising to see basic video prompts as their next release.

With Zoom Out and other tools on the horizon, those generative video prompts are going to keep getting smarter and better release by release.

How do you feel about this Zoom Out tool and the future of AI for video though? Let us know your thoughts in the comments below.