The former Myspace founders create a text-to-video generator that customizes content with your selfie

Chris DeWolfe and Aber Whitcomb, the founders of Myspace and Jam City, have followed tech trends for nearly two decades. The veteran tech entrepreneurs launched Plai Labs, an a16z-backed social platform development startup that provides AI tools for consumers to collaborate and connect, earlier this year.

Plai Labs launched PlaiDay, a free text-to-video generator, today, joining Google’s Imagen, Meta’s Make-A-Video, OpenAI’s DALL-E 2, and Stable Diffusion. However, adding selfies to videos is a standout feature.

PlaiDay lets you watch an animated version of yourself fly over the Grand Canyon or fight intergalactic enemies. Upload a selfie and type a few words to create a short video to share with friends and followers. The videos are three seconds long, but they will get longer. The company is also adding audio.

A company example used the prompt: “English Bobby, 1800s style, streets of London, close-up, life-like.”

PlaiDay’s facial expressions are realistic but not lifelike. Additionally, the background people walk smoothly. They move slightly differently from humans, but it’s not jarring.

The personalised video is sloppy. Since the user’s selfie doesn’t show a mustache, the video appears to show facial hair fading into the skin.

Another example by @Marvel on Discord is 10 times more distorted. Honestly, creepy. Instead, type “close up” or “portrait of” to generate detailed faces, according to the company. PlaiDay can generate one face rather than a group because it’s early.

It may be a while before cinematic AI-generated videos appear on TV. We’re still impressed by PlaiDay’s work and see potential.

The PlaiDay Discord channel offers a personalized text-to-video generator. It will launch on PlaiDay, a free alpha app. PlaiDay also has fun features like inserting yourself into TikTok videos, a text-to-image tool, image resizing, caption generators, and more. The company promises app details in the coming weeks.

“This is just the very beginning of what we see as the future of storytelling,” co-founder and chief architect Jim Benedetto told . “Once you put yourself into these [AI] videos, you tell your own stories.” Benedetto was Myspace’s technology SVP. He co-founded Gravity, which AOL acquired in 2014.

Orchestra, the company’s AI platform, powers the text-to-video generator with in-house and open source models. Plai Labs plans to sell many more products like this.

Orchestra is a “flexible set of tools that can be used to enhance almost any business,” according to the team. It uses reusable code blocks to help tech and non-technical people build and deploy interesting AI applications quickly and efficiently, regardless of industry.

Plai Labs hopes its platform can help designers and product managers build AI without engineers.

“We are excited because it allows a single product manager with a vision to create AI that normally would take five to 10 PHDs,” Benedetto said.

Whitcomb explained that the AI platform goes beyond generative art. Marketing, security, analytics, and more will soon be possible with the platform.

“A diffusion process can generate creatives. Consider analyzing each creative with a neural network and distributing copies. Whitcomb suggested using speech-to-text to index it and an LLM module to use multiple languages or recommend other copy.

“It’s flexible enough to do anything in AI,” he said.

In coming months, the unnamed AI platform will be widely available.