I Used OpenAI’s Sora to Generate Videos of My Cat. Here’s What Happened

OpenAI’s Sora 2 will generate amazing short videos from your text descriptions and uploaded images. But with the latest skill, the AI model is able to create videos based on animals, inanimate objects, and other items around your home or from your own imagination.

Known as Character Cameo, this capability is especially cool if you want to place your pet dog, cat, bird, or hamster in a creative video. It can also capture ordinary household objects and give them life through a cameo video. For this article, I focused on using Sora 2 to create videos of my cat. Here’s how it fared.

To create a Character Cameo, you’ll need a ChatGPT account. A free one will suffice, but a paid Plus or Pro account will grant you more features. You can cook up the video at the Sora website or through the Sora iOS app. You can’t yet create these types of side characters in the Sora Android app, but that support should be on the way. For now, choose the website or iOS app and sign in with your ChatGPT account.

You’re able to control the audience for a particular character. You can set up a video so that it’s viewable and usable only to yourself, to people you approve, to mutuals (people you follow and who follow you), or to everyone. For example, if you set everyone as the audience, that means anyone using Sora can not only see the video but customize it to create their own videos.

Shoot Your Video

To kick things off, you’ll need to shoot a video of the animal or object you want to use for the Character Cameo. Keep the video anywhere from 3 to 10 seconds in length. OpenAI also offers a couple of tips. Keep the character visible for the entire clip. For pets, record your video at eye level and capture natural movement such as walking, sitting, or turning. For objects, hold your phone or camera steady and avoid quick pans.

For this initial test, I shot several videos of my cat, Mr. Giggles. I kept each one under 10 seconds and focused on Mr. Giggles for each clip. I purposedly took multiple videos so that I’d have a few to chose from. The video can be in any one of several formats, including mp4, mov, and avi. Since I shot the video on my iPhone, the clip was automatically formatted as an MOV file.

Upload Your Video

After you’ve shot the video, head to the Sora website or open the Sora mobile app and switch to your profile page. Select the button for Create character. In the iOS app, choose the option to Create side character and then top the button for Upload Video. At the website, click the Upload video button. In the mobile app, you can also record the video on the fly, but I find it more convenient to record the video ahead of time.

(Credit: Lance Whitney / OpenAI)

Select the video you want to use. Here, I chose one of the videos of Mr. Giggles that I thought would work best. After the video appears in Sora, you can trim it if you wish to shorten it or remove any extraneous parts. Move the left or right slider to trim your video. When done, click or tap the right arrow. The video is then uploaded. If it looks good, select Continue. Otherwise, select Retake to try again.

(Credit: Lance Whitney / OpenAI)

Create Your Character

Sora assigns a username and display name to the character, but you can change either of those. Tap one and then the other and type a new name if you wish. The next screen adds a character description, but you can modify this or delete it completely. When finished, select Create character.

(Credit: Lance Whitney / OpenAI)

At the next screen, choose the audience for the character to determine who can see and use it. I typically start off with Only me until I can fine-tune the character. I can then expand it to a wider audience. Select the option for Character preferences to set a pronoun, a description, and any restrictions. When finished, tap or click Done.

(Credit: Lance Whitney / OpenAI)

Cast Your Character in a Video

As the next step, you now cast the character you created in a video. In the mobile app, tap Edit Characters, choose the character to use, and then tap Cast. Alternatively, tap the plus button to create a video. At the prompt, type the @ symbol, start typing the name of the character, and then choose it from the results.

(Credit: Lance Whitney / OpenAI)

At the website, type the @ symbol at the prompt, begin typing the name of the character, and then choose it from the results.

Get Our Best Stories!

Get the Most Out of Your Technology

Sign up for the Tips, Tricks, and How-To newsletter for all of our best expert advice.

Sign up for the Tips, Tricks, and How-To newsletter for all of our best expert advice.

By clicking Sign Me Up, you confirm you are 16+ and agree to our Terms of Use and Privacy
Policy.

Thanks for signing up!

Your subscription has been confirmed. Keep an eye on your inbox!

(Credit: Lance Whitney / OpenAI)

Next, type a description of the video you want to generate with the character. Select the Settings icon to choose the format (portrait or landscape) and the duration for the video. Then click or tap the arrow to submit your prompt.

Using Mr. Giggles as my character, I wrote the following description as suggested by ChatGPT:

“@lancewhit.giggles  A photorealistic cinematic video of Mr. Giggles, a black cat, as an original feline hero. He wears a custom, flight suit with a subtle cape designed uniquely for him. Mr. Giggles glides and soars naturally through open sky above a city at sunset. The camera maintains close, steady tracking with clear, sharp focus on Mr. Giggles’ face and eyes. Natural lighting, ultra-realistic fur texture, lifelike motion physics, grounded realism. No resemblance to any existing superhero characters or franchises.”

You’ll notice that I added that last phrase about no resemblance to any existing superhero characters or franchises. That’s because Sora is very particular about the types of videos you ask it to generate. If it even remotely “thinks” that the content may be similar to copyrighted third-party material, it will refuse to generate the video. (Though it recently signed a deal with Disney to allow Sora videos featuring 200+ characters.)

(Credit: Lance Whitney / OpenAI)

In response, Sora created an amazing video that highlighted Mr. Giggle as a superhero. I could clearly see a medium shot face followed by a close up of one of his eyes as he took to the skies.

(Credit: Lance Whitney / OpenAI)

Next, I wanted to generate a video of Mr. Giggles speaking to me as he asks for his favorite food. Here, I submitted the following prompt:

“@lancewhit.giggles A photorealistic cinematic scene of @lancewhit with his black cat, Mr. Giggles, in a softly lit kitchen or dining room. Mr. Giggles sits at eye level and speaks clearly, with visible mouth movement synced to dialogue. We can see both Mr. Giggles and Lance. Looking at Lance, Mr. Giggles says: ‘I formally request my favorite dinner: tuna tartare, a twice-baked potato, asparagus, and apple pie for dessert.’ He speaks in a soft, feline type voice. Natural lighting, ultra-realistic fur, lifelike facial animation, grounded realism.”

In response, Sora generated a video with Mr. Giggles and me sitting in a kitchen as he announced his dinner request and I responded.

(Credit: Lance Whitney / OpenAI)

Overall, I found Sora a clever and creative way to create character videos. I did have to fiddle with the prompt a couple of times to get the results I wanted. I suggest you ask ChatGPT or another AI to write the prompt for you based on what you want. For me, that helped get past some of the pitfalls that often crop up. Ultimately, both of the videos brought Mr. Giggles to life in ways I hadn’t imagined before.

About Our Expert

Lance Whitney

Contributor

---

Experience

I’ve been working for PCMag since early 2016 writing tutorials, how-to pieces, and other articles on consumer technology. Beyond PCMag, I’ve written news stories and tutorials for a variety of other websites and publications, including CNET, ZDNet, TechRepublic, Macworld, PC World, Time, US News & World Report, and AARP Magazine. I spent seven years writing breaking news for CNET as one of the site’s East Coast reporters. I’ve also written two books for Wiley & Sons—Windows 8: Five Minutes at a Time and Teach Yourself Visually LinkedIn.

I’ve used Windows, Office, and other Microsoft products for years so I’m well versed in that world. I also know the Mac quite well. I’m always working with iOS, iPadOS, watchOS, and Android on my various mobile devices. And these days, I write a lot about AI, so that’s become another key area for me.

My wife always jokes about all the tech products we have around the house, but I manage to put them to good use for my articles. I like Lenovo computers, so I own a couple of Lenovo desktops and several laptops. I have three MacBooks and a Mac mini. For my mobile life and work, I use an iPhone 16 Pro, iPad Pro, and iPad mini as well as an Apple Watch. But since I write about Android, I own several Android phones and tablets. Like any tech person, I have a cabinet full of cables, wires, and assorted mysterious gadgets. And when it’s time to take a break from writing, I have an old Xbox 360 and Nintendo Wii, both of which I use for exercise and fitness games.

Read Full Bio