Seattle-area startup Irreverent Labs has shifted its focus from blockchain-based video games and NFTs to artificial intelligence. Specifically, it wants to build foundation models for text-to-video generation and related content creation tools. Text-to-video is being explored by several companies but is still in development. Samsung Next was intrigued enough with the proposition to invest an undisclosed sum in Irreverent. While there are several apps that output cartoonish results, ambitious efforts are limited. Animations that aim for photorealism, such as Meta’s Make-a-Video and Runway’s Gen-2, can output only four or five seconds of video at a time.

Creatives in film, television and gaming are using popular AI image generators like Midjourney, Dall-E 2 and Firefly to make still images that can be processed through software like Blender and After Effects to create video. Popular use cases include previsualization and LED backgrounds in virtual production.

The goal of using words to prompt high-quality video of substantive length remains elusive. But AI is advancing rapidly, and a group of companies that now includes Irreverent Labs sees it within reach.

Founded in 2021, Irreverent Labs raised $45 million last year in a round led by Andreessen Horowitz. At that time, Irreverent was best known as a blockchain company that developed a game around NFTs, “MechaFightClub.”

“But the company insists now that the game was a way to showcase what is essentially a large machine learning model that … will allow users to make videos using various inputs, from images to text to audio, later this year,” according to TechCrunch.

“Irreverent Labs said its upcoming ‘video foundation model’ can seamlessly translate prompts into captivating 3D videos with an unprecedented level of quality and control,” writes VentureBeat, noting that “experts say the technology could significantly lower barriers for producing video entertainment” for experts and amateurs alike.

“The opportunity to be embedded on Samsung devices is pretty significant for us,” co-founder and CEO Rahul Sood told VentureBeat. Sood met Irreverent Labs co-founder and CTO David Raskino in 2011 while the two were working at Microsoft.

“Making it easy for people to take a series of photos and submit them to [our foundation model] to create a video is pretty interesting — and it will open up a ton of use cases for us that we have not yet discovered,” Sood further explained to VentureBeat.

Irreverent’s immediate focus will be on entertainment professionals in gaming, film and television. It hopes to release an API as part of a developer preview later this year.

Related:
Why We Invested in Irreverent Labs, the Future of High-Fidelity Video Creation, Samsung Next, 8/24/23