Will DePue, who helped create OpenAI’s Sora model, returns to lead a new team with Troy and Eric Luhman focused on the pursuit of artificial superintelligence.
OpenAI engineer Will DePue, one of the key developers behind the company’s text-to-video model Sora, has announced his return to the firm to lead a new research team dedicated to pursuing Artificial Superintelligence (ASI).
In a post on X on Wednesday October 29, 2025, DePue said he is forming the team alongside Troy Luhman and Eric Luhman, both fellow Sora engineers. The group’s mission, he wrote, is to take “an incredibly high-risk bet that has a small but significant chance of leading to ASI.”
“We’re keeping the team tight,” DePue added, “but we’re open to high-slope researchers and engineers interested in making progress in uncharted territory.”
OpenAI has not yet released details about the new initiative, and the company did not immediately respond to a request for comment.
DePue, born in 2003, studied Computer Science at the University of Michigan before leaving to focus on research and development projects. He first joined OpenAI in mid-2023, where he worked on training ChatGPT models and played a central role in developing Sora — the company’s groundbreaking text-to-video generator unveiled in February last year. Sora was publicly launched in December, followed by an upgraded version, Sora 2, in September.
Read Also: Zuckerberg’s Meta $70 Billion AI Drive Sparks Market Concerns
Beyond his work at OpenAI, DePue has published several experimental projects on his website, including a browser-based GPT implementation, a Turing-complete computer built within Figma, and a dataset project titled the “Alexandria Index.”
Troy and Eric Luhman, who are joining him on the new team, have co-authored several research papers in machine learning but remain relatively private figures in the AI community.
The announcement comes as OpenAI continues its broader mission to achieve Artificial General Intelligence (AGI) — the stage where AI can reason and learn at human levels. However, CEO Sam Altman has recently emphasized that the company’s ambitions now extend beyond AGI.
In a January blog post, Altman wrote that OpenAI is “looking ahead to superintelligence,” a stage where AI systems could think, create, and reason far beyond human capability. Such advancements, he noted, could “massively accelerate scientific discovery and innovation” and “increase abundance and prosperity.”
DePue’s new initiative reflects that long-term vision — one that seeks to explore the outer limits of what intelligent systems might someday become.