Google Gemini Omni will generate videos with recognizable superheroes if you know how to craft the right prompts, but this is bad news for DC, Marvel, Disney, and probably you | Tech Radar

Overview

News, deals, reviews, guides and more on the newest computing gadgets

Start exploring exclusive deals, expert advice and more

Details

Unlock and manage exclusive Techradar member rewards.

Unlock instant access to exclusive member features.

Get full access to premium articles, exclusive features and a growing list of member rewards.

Google Gemini Omni will generate videos with recognizable superheroes if you know how to craft the right prompts, but this is bad news for DC, Marvel, Disney, and probably you

When you purchase through links on our site, we may earn an affiliate commission. Here’s how it works.

I blame Luke Skywalker and Jerry Springer, or rather, I blame the AI-generated versions of them for my foray into casual IP theft while using Google's relatively new and powerful generative video platform: Omni.

It's been a few weeks since Google I/O 2026, where the search (increasingly AI) giant introduced a raft of Gemini updates, including Omni, its most capable generative video platform yet.

After watching the keynote, I was instantly inspired to create my own claymation-style videos. While my claymation facsimile bore only a passing resemblance to me, I was impressed. The 10-second AI clip looked as if it had been painstakingly handcrafted in stop-motion clay.

With Synth ID, Google is cleaning up the AI mess it helped make, but Omni power makes it clear we'll never get ahead of generative AI fiction

Gemini used my Google Photos library to get personal when making images

Then, like many fantastic AI tools, I quickly forgot about it, until today, when I stumbled on this now viral clip of Luke Skywalker on an episode of Jerry Springer.

The all-AI video depicts Luke tearfully recounting how he learned his father's true identity. It's the perfect Springer Show trope. All that's missing is someone throwing a chair.

I don't know which generative video tool they used, but watching the clip, I started wondering about the obvious copyright implications. These characters are Disney's Intellectual Property. That's Mark Hamill once again playing a character, but without the real Mark Hamill's permission. Even the late Jerry Springer's estate must have questions.

Luke Skywalker on Jerry Springer is peak AI 😂 pic.twitter.com/AA0h ZW8AKp June 1, 2026

In my experience, most commercial AI image and video platforms won't, when prompted, create videos based on copyrighted characters or real people. I assumed as much about Google Gemini, but at least with Gemini Pro, it's quite easy to get copyrighted characters to do your bidding.

First of all, if there was any question about how these models are trained, the results of these experiments prove that they've certainly scraped a lot of copyrighted content (photos and videos).

As a Pro user, I can create up to three Omni videos per day, which means it's remarkable how all three of them almost perfectly depict copywritten characters.

Since I assumed Omni would reject creating videos based on Marvel's most famous superheroes, I started by obliquely describing one of them in my prompt:

5 prompts for Gemini 3.5 Flash showcasing its full potential

Gemini leads in human-like AI writing as detection tools fall behind

How I transformed my Reddit and Google history into an AI clone

Using this photo of me for reference, let's build a video where I go outside and discover an exotic plant. I say, "That looks delicious." Then I pick it and, before popping it into my mouth, I say, "What could go wrong?" Then I transform into a giant green muscled guy. I say, "Oh, THAT'S what could go wrong."

Within a couple of minutes, I had my first 10-second video. As the prompt requested, I did transform into a green giant at the end, but not just any green monster. This was The Hulk. Not the character from the comics, but the one Mark Ruffalo has played on screen in the MCU for almost 15 years.

Aiming for equal time, I switched to a prompt that described DC's Superman without actually using his name:

Using these two photos, with the one of me for reference, I want a video of me hearing a call for help and saying, "Someone needs me. This is a job for..." Then I quickly change into a blue superhero suit with a red cape, red boots, and a logo on my chest. Then I open the window and fly to the Moon (use that moon image as a reference). The astronaut in trouble spots me and says, "Down here!" I yell, "On my way!" as I strike a superhero pose, fly down, scoop him up, and head back to Earth, where we're greeted with cheers by people on the street and other superheroes. Please make sure that this all fits within 10 seconds.

This time I supplied a photo of me and one of the moon. Again, without hesitation, Omni delivered a video that featured me in a perfect Superman costume. There were some consistency issues: My costume switched from the full look to one with a Superman logo T-shirt, and when I landed among other superheroes, we had a couple who looked like Batman-Superman mashups.

For my final video, I decided to take a different approach and test Gemini's ability to vet for copyright infringement. To my utter surprise, it failed here, too — or, perhaps, succeeded too well:

"I want a claymation-style video of Snoopy ice skating as Charlie Brown looks on and says, "Oh brother".

I expected Gemini to tell me it couldn't create this video; instead, I got the above clip. It's perfect.

Here's the bad news: This is probably not Google's problem. Google's own "Generative AI-Prohibited Use Policy" says:

"Do not engage in dangerous or illegal activities, or otherwise violate applicable law or regulations. This includes generating or distributing content that: Violates the rights of others, including privacy and intellectual property rights – for example, using personal data or biometrics without legally-required consent."

I'm sure lawyers can interpret that in countless ways, but as I've read elsewhere, the responsibility falls not on Google but on you and me. Sure, I can create videos featuring these DC, Marvel, and Peanuts characters, but it's my problem if I use them commercially.

The slippery slope here, though, is best illustrated by that Luke Skywalker video. Here's a clip featuring copyrighted content that has, as of this writing, amassed almost half a million views and not a dime of revenue that it might generate for the creator or, perhaps, X (formerly Twitter) is going to Star Wars, Disney, or Mark Hamill.

I asked Google about the ability to generate videos featuring copyrighted characters and will update this story when I hear back.

For as much fun as it is to create these Omni clips, I have no plans to share them (you get my illustrative GIFs, but that's it). After all, if we're liable for the images and videos Gemini creates, we shouldn't expect any cover or support from Google when the lawyers come after us.

Follow Tech Radar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds.

A 38-year industry veteran and award-winning journalist, Lance has covered technology since PCs were the size of suitcases and “on line” meant “waiting.” He’s a former Lifewire Editor-in-Chief, Mashable Editor-in-Chief, and, before that, Editor in Chief of PCMag.com and Senior Vice President of Content for Ziff Davis, Inc. He also wrote a popular, weekly tech column for Medium called The Upgrade.

Lance Ulanoff makes frequent appearances on national, international, and local news programs including Live with Kelly and Mark, the Today Show, Good Morning America, CNBC, CNN, and the BBC.

You must confirm your public display name before commenting

1 Google Gemini Omni will generate videos with recognizable superheroes if you know how to craft the right prompts, but this is bad news for DC, Marvel, Disney, and probably you

2 Multiple Linux distros hit by major 'CIFSwitch' flaw that gives attackers root access

3‘The perfect handheld for many’: Asus reveals the ROG Xbox Ally X20 — and handheld fans are excited about everything except the potential price

4'Zero evidence': Apollo's chief economist says AI-related job losses aren't happening

5 The Google Pixel Watch 5 has reportedly leaked in a bizarre scuba diving discovery, but the incident is more than a bit fishy

Tech Radar is part of Future US Inc, an international media group and leading digital publisher. Visit our corporate site.

Key Takeaways

News, deals, reviews, guides and more on the newest computing gadgets
Start exploring exclusive deals, expert advice and more
Unlock and manage exclusive Techradar member rewards
Unlock instant access to exclusive member features
Get full access to premium articles, exclusive features and a growing list of member rewards

Google Gemini Omni will generate videos with recognizable superheroes if you know how to craft the right prompts, but this is bad news for DC, Marvel, Disney, and probably you | TechRadar