AI video technology startup Runway doesn’t have the everyday Silicon Valley pedigree. No Stanford founders, no ex-Google founders, no nine-figure seed spherical that purchased them time to forget about earnings. Its 3 founders — two from Chile, one from Greece — met at NYU’s Tisch College of the Arts and constructed the corporate in New York.
Runway additionally may well be, relying on who you ask, some of the consequential AI firms as of late. No longer as a result of what it has constructed, however as a result of what it is making an attempt to construct subsequent.
For the previous a number of years, the AI business has in large part operated at the premise that intelligence lives in language. Massive language fashions like OpenAI’s ChatGPT and Anthropic’s Claude mirror that wager.
Runway, along different competition, is making a special one. Its founders imagine the following type of AI intelligence received’t be constructed from textual content, however from video and international fashions that learn the way the sector works, now not simply how people describe it. That difference sounds educational. Its implications aren’t.
Runway co-Founder and co-CEO Anastasis Germanidis stated coaching fashions at once on observational knowledge from the sector is the following frontier of AI. The firms that get there first, he argues, received’t be those who perfected language.
“We’re principally sure by means of our personal working out of fact,” Germanidis instructed TechCrunch from Runway’s homey sunlight-filled headquarters close to Union Sq..
“Language fashions are skilled on all the web, on message forums and social media, on textbooks — distilling the prevailing human wisdom,” Germanidis persevered. “However to get past that, we wish to leverage much less biased knowledge.”
Based in 2018, Runway constructed its popularity on video-generation fashions — together with its newest Gen-4.5 — and AI gear that permit other people flip textual content activates into editable, cinematic content material.
These days, Runway’s era powers manufacturing workflows for filmmakers and advert companies, and the corporate has signed offers with primary media avid gamers like Lionsgate and AMC Networks. Its gear have even been utilized in movies corresponding to “The whole lot All over All At As soon as.”
Runway is now valued at $5.3 billion and, in step with considered one of its founders, added $40 million in annual habitual earnings in the second one quarter of 2026.
If Runway’s wager that video technology is the trail to international fashions can pay off, the end result shall be felt from Hollywood to drug discovery. If it doesn’t, Runway dangers being outpaced by means of competition with a long way deeper wallet — Google leader amongst them.
Taking the bounce
Inside the ultimate six months, the startup has put its plan into motion and expanded past video technology, launching its first international type in December, with plans to release some other this yr. (Global fashions are AI methods that simulate environments properly sufficient to expect how they’ll behave.)
Runway isn’t on my own in its pursuit of turning physics-aware video fashions to international fashions, with near-term use circumstances in interactive leisure, gaming, and robotics coaching. Startups Luma and Global Labs are on a an identical trajectory, and Google has pointed its Genie international type in the similar course.
Everyone seems to be after some model of the similar factor: AI that solves humanity’s toughest issues. That’s a long way from Runway’s authentic product, however it’s the results of each emergent functions within the era and founders who have been predisposed to practice the place it led.
For his section, Germanidis sees international fashions as clinical infrastructure. The extra sensory knowledge and observations you teach a unmarried type on, the nearer you get to a operating virtual dual of the universe — one you’ll run experiments on sooner than any lab may. A lot of the clinical procedure is simply ready on effects, he issues out. If you might want to compress that ready, you might want to compress growth itself.
“If we will be able to construct a greater scientist than human scientists, we will be able to boost up growth in how we perceive the universe and the way we clear up issues,” Germanidis stated.
The moonshot
Germanidis fell in love with programming as an 11-year-old in Athens and got here to the U.S. at 18 to check neuroscience and movie. He grew to become again to laptop science, operating at a number of Silicon Valley tech corporations earlier than deciding he’d had sufficient of the tradition. Co-CEO Cristóbal Valenzuela, born and raised in Santiago, studied economics as an undergraduate earlier than operating in movie after which instrument. Any other Santiago local, Leader Innovation Officer Alejandro Matamala-Ortiz studied promoting and ran a design company.
The 3 met in 2016 whilst attending NYU’s ITP (Interactive Communications Program), a graduate program that Valenzuela described as an “artwork college for engineers.”
The co-founders had all aspired to be filmmakers at positive issues of their lives, in step with Matamala-Ortiz. So Runway began with a easy venture: Are we able to use AI to make everybody a filmmaker?
After liberating their first video technology type in February 2023 — which is staggeringly unimpressive in comparison to what Runway is hanging out as of late — that venture advanced into: May we make everybody a nice filmmaker, in step with Matamala-Ortiz.
It required rising the staff to what it’s as of late. The corporate has 155 staff unfold throughout workplaces in New York, London, San Francisco, Seattle, Tel Aviv, and maximum not too long ago, Tokyo. “However all the way through this procedure, we realized that those fashions can know the way the sector works, and in case you scale them, they may be able to be helpful for lots of different various things,” he added.
Such things as robotics, drug discovery, and local weather modeling — the sorts of issues that experience stumped researchers for many years. Final yr, Runway introduced a robotics unit which Germanidis says has already led to real-world trying out and deployments.
Germanidis, like others, sees the sphere heading towards coaching a unmarried type on many alternative modalities — textual content, video, voice, and different sensors — and thinks the compounding impact is the purpose.
His personal moonshot purpose for Runway’s era, given sufficient time and assets, is organic international fashions and anti-aging analysis.
Whether or not Runway can raise its video dominance into international fashions is a long way from settled, and the contest isn’t ready round. Runway was once a few of the first to AI video technology, however international fashions are a special race with deep-pocketed and well-respected competition. Google, former Meta leader scientist Yann LeCun, AI’s ‘godmother’ Fei-Fei Li, and a rising box of startups are all chasing the similar purpose.
Kian Katanforoosh, CEO of AI talents benchmarking corporate Workera and a lecturer at Stanford, identified that nobody has but confirmed the bounce between video intelligence and generalized reasoning by the use of international fashions, however that doesn’t imply it’s unattainable. He stated that if Runway needs to show its international type wager into fact, it’ll wish to proceed accumulating assets — compute leader amongst them.
Runway has offers with CoreWeave and Nvidia, however wouldn’t verify whether or not it has devoted cluster get admission to — the type of assured, large-scale compute that coaching frontier fashions calls for.
“How can you construct a foundational type and not using a cluster?” Katanforoosh requested. “I don’t assume anyone can do this.”
Runway has raised $860 million thus far, together with a $315 million spherical in February from strategic companions like AMD Ventures and Nvidia. That’s kind of consistent with its maximum speedy competition, Luma AI and Global Labs, that have raised $900 million and $1.29 billion, respectively, in step with PitchBook.
However Runway could also be going up towards incumbents like OpenAI, which has raised round $175 billion in keeping with CEO Sam Altman, and tech behemoth Google, whose mother or father corporate Alphabet is price $4.86 trillion. Google is Runway’s largest danger. The corporate’s Veo type competes at once with Runway’s video technology industry, whilst its Genie international type objectives the similar longer-term territory Runway is racing against.
Katanforoosh nodded at OpenAI, which shuttered its video platform Sora in March after burning kind of $1 million in keeping with day in compute prices with slightly $2.1 million in earnings in accordance to a couple estimates. His level: assets on my own don’t ensure survival. They don’t ensure it for Runway both.
Katanforoosh isn’t writing Runway off. He pointed to AI audio startup ElevenLabs, which has outperformed OpenAI and Google on their very own benchmarks, in spite of missing the assets and pedigree of both. Runway, he argues, may practice a an identical playbook.
The comparability isn’t misplaced on Runway’s founders. Valenzuela says the startup’s loss of Bay Space “standardization” offers them an edge. No longer best do they’ve variety of idea, he contends, however with out Silicon Valley ties, they needed to be scrappier, missing the battle chest many in their friends have get admission to to that will have insulated them from the wish to generate earnings early.
And in step with Michelle Kwon, Runway’s leader running officer, the corporate isn’t in a hurry to boost extra price range, at the same time as compute calls for building up with scale.
“Their background has led them to be early, to be proper extra ceaselessly than now not, and to construct a tradition that strikes extremely temporarily,” early investor Michael Dempsey, managing spouse at Compound, instructed TechCrunch.
For Valenzuela, that tradition begins with how he sees the sector within the first position. He spends no matter unfastened time he has — now not a lot, as a co-CEO and new father — studying books, together with the Chilean poet Nicanor Parra, whom he describes because the antithesis of Pablo Neruda: much less formal, much less educational, maintaining a view that poetry belongs to the folk fairly than to laws.
“Regulations are simply laws they invented,” Valenzuela stated. “That’s a motive force of the way we do issues at Runway. They are saying Silicon Valley is right here and that’s the place the startups are. Why? The ones are simply made up laws. Scrub all of them and get started once more.”
While you acquire via hyperlinks in our articles, we might earn a small fee. This doesn’t impact our editorial independence.
