Runway draws fresh $141 million as next-level generative AI video begins to emerge

Runway, one of the hottest generative AI startups with its text-to-image video tools, has announced a fresh round of funding, adding $141 million in a series C from Google, Nvidia and Salesforce Ventures, among other investors.

The New York City-based company said in a press release that it will use this new financing to “further scale in-house research efforts, expand its world-class team, and continue to bring state-of-the-art multi-modal AI systems to market, while building groundbreaking and intuitive product experiences.”

Runway began with a mission to build AI for creatives

In March, VentureBeat spoke to Runway CEO and cofounder Cristóbal Valenzuela. He discussed the gated launch of Runway’s Gen-2 tool, which is now generally available, and the company’s founding four years ago with a mission to build AI tools specifically for artists and creatives.

“Since then, we’ve been pushing the boundaries of the field, and building products on top of that research,” he said, saying Gen-2 is a “big step forward” in the company’s text-to-video efforts. He pointed to the company’s millions of users, ranging from award-winning movie directors and advertising and production companies all the way down to small creators and consumers.

“We’ve built an incredibly tight community that has helped us understand how actually creatives are using generative AI in their work today,” he said, pointing to Runway’s work for the Oscar-winning movie Everything Everywhere All at Once. One of the film’s editors used Runway to help with effects on a few shots.

“So we have a lot of folks who have helped us understand how these models are going to be used in the context of storytelling,” he explained. “We’re heading to a world where most of the content and media and videos that you consume will be generated, which requires a different type of software and tools to allow you to generate those kind of stories.”

Runway’s growth comes as artists push back on generative AI

Runway’s efforts, however, come at a time when artists are pushing back against generative AI. For example, thousands of screenwriters have been on strike for over two months, halting many movie and television productions, because they want limits on the use of generative AI.

And VentureBeat recently reported that Adobe Stock creators are unhappy with the company’s generative AI model Firefly. According to some creators, several of whom VentureBeat spoke to on the record, Adobe trained Firefly on their stock images without express notification or consent.

There are also several lawsuits pending in the generative AI space. Just today, for example, plaintiffs filed suit against OpenAI, claiming the company used “stolen data” to “train and develop” its products including ChatGPT 3.5, ChatGPT 4, DALL-E and VALL-E. 

Three cofounders attended art school

“We do a lot of listening and are part of the community,” said Valenzuela, pointing to Runway’s AI Film Festival in March as an example of driving conversations and understanding how these technologies will be used by professional filmmakers and storytellers.

“I do think there’s confusion around how these algorithms are already being used in creative environments,” he said. “There’s a misconception that … you have the systems do everything for you and you have no input. We don’t see it like that. We see these tools as tools for human augmentation. They’re tools for enhancing creativity. They’re not tools for replacing creativity.”

Valenzuela emphasized that he comes from an art background. “I went to art school and I started Runway while I was an artist,” he said. “These are tools I wanted to use.”

Originally from Chile, Valenzuela came to New York City to attend the Tisch School of the Arts at New York University — where he met his cofounders Anastasis Germanidis and Alejandro Matamala — but soon realized that his artwork was better suited to making tools.

“My art was toolmaking, I was eager to see artists using the tools I was making,” he said. “So I went deep into the rabbit hole of neural networks — the idea of computational creativity.”

As far as commenting on issues of copyright, fair use and work replacement cited by artists, Valenzuela maintained that it is still “very early” in understanding all the implications of generative AI. “We’re really trying to make sure we can drive this conversation to a positive end,” he said. “I think listening is still the most important aspect. I think being open to change and being able to adapt and understand how things are going to be used, those are the driving factors of how we think about our product. I can’t really speak for other companies and how the other companies are thinking about the space, but for us, we have the commitment to our users.”