Generative artificial intelligence, used to create content from text to images to videos, is still on the rise. Big Tech and startups across the globe are racing to compete and launch their own AI-powered chatbots, text-to-image generators, and video tools.
Among them is a Bellevue, Washington–based startup called Irreverent Labs, whose AI technology enables anyone to create a video, and which, like a lot of its rivals, just closed on a new round of funding. Samsung Next led the round; Irreverent isn't disclosing how much due to Samsung's corporate policy.
The round is not the first for Irreverent Labs, which was founded in 2021 and raised $45 million in funding last year, led by Andreessen Horowitz. At the time, Irreverent was described by media outlets as a blockchain company, one that had developed a robot cockfighting game called MechaFightClub that used non-fungible tokens (NFTs). But the company insists now that the game was a way to showcase what is essentially a large machine learning model that, in the case of Irreverant Labs, will allow users to make videos using various inputs, from images to text to audio, later this year.
Indeed, according to Irreverent Labs' co-founder and CEO, Rahul Sood, the company elected to work more closely with Samsung as a strategic investor partly to access Samsung Next portfolio companies that may want to use its API. Further, the outfit will be working with Samsung's device units in order to develop a larger distribution strategy, Sood explained.
Sood, a serial entrepreneur, founded Irreverent Labs with David Raskino, who today is the company's CTO. The duo met in 2011 at Microsoft, where they set up Microsoft Venture together. Sood left Microsoft for a new game venture, while Raskino started a venture fund in 2014.
The co-founders' conversations about Irreverent gained momentum after Sood's previous company was acquired in 2021. "While David aspired to construct an AI application for business purposes, I was more interested in doing something around entertainment," Sood said. "Consequently, the inception of Irreverent Labs merged, characterized as an 'automated entertainment company.'"
Though it initially used its AI technology to build video games, it realized that allowing users to generate all types of short videos, including 3D, presented a much bigger opportunity.
"At Irreverent, we're on a mission to make it possible for anyone to create compelling short-form video entertainment with AI," said Sood, explaining how the product would work. "Users can visit the website and upload a short video prompt like a live photo. We will then predict a short, high-quality video from these frames. It's super simple." (Irreverent will announce the name of the website users can visit later this year.)
The startup will eventually make it possible to use multiple inputs like text to direct the action and add characters, audio, and the beginning/ending frames users can upload, according to Sood. "We'll generate the movie to fill in the blanks," he added.
Initially, Irreverent's users will likely be people who are already making short-form content regularly and uploading on platforms like Tiktok, YouTube and Instagram. Later, it expects its AI foundation models will enable "people who do not have classical video production experience to create short video content." The company will also be targeting the developer ecosystem, as it plans to release an API, as well as enterprise customers.
"The possibilities that Irreverent Labs' technology unlocks are vast, and the potential impact on the mobile devices in our pockets and backpacks and the televisions mounted on our living room walls are immense," said Joan Kim, investor at Samsung Next. "From concept to reality, this revolutionary model bridges the gap between imagination and execution."