Origin Lab Raises $8M to License Video Game Data for AI World Models

Origin Lab raised $8 million in a seed funding round led by Lightspeed Ventures. SV Angel, Eniac, Seven Stars and FPV participated in the round, which also included angel funding from Kevin Lin, co-founder of Twitch, and Kyle Vogt, founder of Cruise. Anne-Margot Rodde, co-CEO and co-founder of Origin Lab, said the company addresses a critical gap for AI development.

“The AI systems that are being built now need to understand how the physical world works and how things move,” Rodde said. ” Origin Lab serves as a marketplace where world-model-focused labs can buy high-quality licensed data from video game companies.

The startup converts video game assets into training data for world models, ranging from simple rendering runs to automated hours of walkthrough footage.

Rodde, along with co-founders Antoine Gargot and Colin Carrier, identified an opportunity in the video game industry’s untapped resources. “It became clear that the video game industry was sitting on some incredibly valuable data, but there was no real way or infrastructure to connect AI labs and the video game industry.

The timing reflects growing demand. In December 2024, OpenAI’s first version of its Sora video-generation model seemed to regurgitate footage of popular video games and streamers. Amazon has been open about its interest in using Twitch footage to train models.

Faraz Fatemi, a partner at Lightspeed who led the Origin investment, pointed to strong market signals from existing data vendors. “We’ve seen how sharp the revenue scaling can be for data vendors that are serving the major labs,” Fatemi said. ” Separately, a career development platform went on sale.

99, regularly $297. The platform uses a multi-domain IQ assessment, a Big Five personality profile, and a Career DNA report. IQ Career Lab includes a live database of more than a million job listings.

The lifetime subscription provides unlimited access to assessments and reports, verified IQ and personality certificates, detailed score breakdowns, and personalized career insights. The two announcements illustrate parallel trends in the AI-adjacent economy: one supplying raw data to frontier model builders, the other equipping individuals with psychometric tools to navigate resulting labor shifts.