Luma just launched an AI production studio and its first show stars Ben Kingsley

Luma AI just moved from building tools to making movies. The $4 billion AI video startup has launched a full production company called Innovative Dreams in partnership with Wonder Project, a faith-and-values streaming brand on Amazon Prime. The first show stars Ben Kingsley. And it’s about Moses. If that sounds like an unlikely pivot for a Silicon Valley AI company, that’s because it is and it tells you a lot about where the AI-in-Hollywood race is actually heading.

Key Takeaways

• Luma AI launched Innovative Dreams, a new production company partnering with Wonder Project (Amazon Prime’s faith-content studio)
• First project: “The Old Stories: Moses” starring Ben Kingsley, arriving on Prime Video this spring
• Uses Luma’s newly released AI Agents for real-time changes to sets, lighting, props, and actor appearance during production
• Director Jon Erwin’s “real-time hybrid filmmaking” merges performance capture and virtual production at significantly lower cost
• Luma is now valued at $4B+ after raising $900M Series C in November 2025 led by Saudi AI firm HUMAIN
• It’s unclear whether Innovative Dreams will remain exclusively faith-focused or expand to other genres

What Luma Just Announced

On April 16, 2026, Luma AI announced the launch of Innovative Dreams, a new production services company created in direct partnership with Wonder Project, the streaming operation run by director Jon Erwin and former Netflix executive Kelly Hoogstraten. The venture is squarely aimed at the faith and values audience a segment of the viewing market that is large, loyal, and historically underserved by the major Hollywood studios.

The first project coming out of Innovative Dreams will be “The Old Stories: Moses,” featuring British actor and Academy Award winner Ben Kingsley. The series is set to drop on Amazon Prime Video this spring. That alone is a remarkable debut for a production company that technically didn’t exist until this week.

In a social media post announcing the venture, Luma described Innovative Dreams as “a production services company where seasoned filmmakers from Director Jon Erwin’s team and Luma’s creative technologists work with great studios and filmmakers to help them realize ambitious ideas.” That framing technologists working alongside established Hollywood professionals is meaningful. It signals that Luma isn’t trying to replace filmmakers. It’s trying to put significantly more powerful tools in their hands.

Real-Time Hybrid Filmmaking What That Actually Means

The technical approach behind Innovative Dreams is what makes this story genuinely interesting. Director Jon Erwin, in a video promoting the partnership, described a new methodology called “real-time hybrid filmmaking.” It combines two of the most expensive techniques in modern cinema performance capture (the technology used in “Avatar”) and virtual production (the LED-screen approach made famous by “The Mandalorian”) but does so live and at a fraction of the traditional cost.

Here’s what that looks like in practice. Traditional performance capture requires an elaborate studio setup where actors wear full-body suits and facial markers. The captured data is then handed off to post-production teams who turn it into animated characters weeks or months later. Virtual production, meanwhile, requires massive LED display stages and real-time game-engine rendering infrastructure that costs millions just to set up. Combining the two typically means an army of specialists working in sequential stages across a months-long pipeline.

Luma’s tools, according to Erwin, collapse that pipeline. Creative teams can work directly with Luma Agents the company’s recently launched AI tools designed for end-to-end creative work across text, image, video, and audio to make real-time changes to sets, lighting, props, and even actor appearance during the shoot itself. As Luma put it: “This is a significant improvement over the current virtual production and performance capture processes where things come together only in post.”

Even more striking is the capability Erwin described for character replacement: film a human actor anywhere, drop that performance into a photorealistic AI-generated scene, and if needed, generate an entirely new face that maps onto the actor’s movements and expressions. The underlying performance the emotion, the physicality, the timing stays human. The visual execution becomes something far more flexible.

Why Wonder Project? The Faith Audience and AI’s Content Opportunity

Wonder Project launched in 2023 with a clear and focused mandate: produce high-quality film and TV content for the faith and values audience, globally. Hoogstraten, who previously led streaming initiatives at Netflix, brought institutional knowledge of what it takes to build a dedicated subscriber base around a specific cultural identity. And Erwin, who has spent his career making faith-oriented films, brought the creative credibility.

The studio’s first project, “House of David” a Biblical drama about the life of King David landed on Amazon Prime in 2025 and reportedly performed well with its target demographic. “The Old Stories: Moses” is the natural follow-up, and the stakes have been raised considerably with Kingsley headlining.

What makes Wonder Project an ideal first partner for Innovative Dreams isn’t just the subject matter it’s the economics. Faith-based audiences are notoriously committed viewers. They will watch a series because it speaks to their values, not simply because it has the biggest special effects budget. That means a production that uses AI to reduce costs while maintaining visual quality is actually a better product for this audience than a bloated, expensive spectacle would be. Luma’s technology makes the most sense where storytelling matters more than spectacle and Wonder Project is exactly that kind of studio.

It’s also worth noting that AI-powered production tools could unlock a wave of faith-based content that previously couldn’t get made at all. Biblical epics are expensive to produce. The sets, the costumes, the scale required to depict ancient civilizations none of that comes cheap. If Luma’s tools can deliver the visual scope of “The Ten Commandments” at the cost of an indie drama, the addressable market for this kind of content expands dramatically. That’s a real opportunity, and it’s one Hollywood has largely left on the table.

Luma Isn’t the Only One Making This Move

The launch of Innovative Dreams arrives in the middle of what looks increasingly like a coordinated industry shift AI video startups moving from providing tools to actually producing content. It’s a logical progression, and Luma is far from alone.

Higgsfield, another AI video startup, launched an original series last week, beginning with a 10-minute sci-fi episode. London-based Wonder Studios — which raised $12 million backed by OpenAI is currently developing a documentary with Campfire Studios. The pattern is clear: the companies building the most capable AI production tools are deciding the best way to prove those tools work is to use them to make something real.

This trend also connects to broader shifts at the studio level. Amazon has been turning to AI to cut production costs for both film and TV projects. Sony Pictures has announced plans to incorporate generative AI into its production workflows. Even James Cameron one of the most technically demanding directors in Hollywood history has come out publicly in support of AI as a tool for keeping ambitious films in production without mass layoffs.

And then there’s the upcoming “Bitcoin: Killing Satoshi,” which is set to become the first studio-quality AI feature film on the market with a reported production budget of around $70 million down from an estimated $300 million without AI involvement, according to TheWrap. That’s not an incremental cost saving. That’s a fundamental restructuring of what’s financially possible in filmmaking.

The Runway Argument and Why It’s More Complicated Than It Sounds

The same week Luma launched Innovative Dreams, Runway co-founder and co-CEO Cristóbal Valenzuela made headlines with a provocative argument at the Semafor World Economy summit. Runway, which is now valued at more than $5 billion, has been aggressively courting Hollywood partnerships for years.

Valenzuela’s pitch was simple: instead of spending $100 million on a single feature film, studios should spread that budget across 50 films using AI. Same total spend. More content. Better odds of a hit.

The creative community predictably pushed back. At SXSW 2026, Steven Spielberg made clear he has no interest in AI as a creative tool and argued that the human element is irreplaceable in cinema. The counterargument is that Valenzuela isn’t talking about replacing human creativity he’s talking about changing the economic model around it. Those are genuinely different conversations, and conflating them doesn’t serve anyone.

What’s interesting about Luma’s approach with Innovative Dreams, compared to Runway’s volume-over-everything pitch, is that it’s more targeted. Luma isn’t saying “make 50 Moses films.” It’s saying “make one Moses film that’s better than what would have been possible before, at a cost that actually makes sense for a faith-focused streaming studio.” That’s a more defensible argument and arguably a smarter one for building long-term credibility in Hollywood. We’ve seen enough examples of companies rushing into AI without a clear creative vision only to regret it. Luma seems to understand that.

About Luma AI — From 3D Capture to Hollywood’s New Toolmaker

Luma AI was founded in 2021 in Palo Alto by CEO Amit Jain, who previously studied machine learning at Stanford and worked on perception systems at Apple. The company’s early work focused on 3D capture and Gaussian splatting technology that allows a smartphone to create detailed 3D models of physical spaces. That foundation in understanding physical reality, and training AI to replicate it, became the technical core of everything that followed.

The company’s Dream Machine platform, which generates professional-grade video from text prompts, now has more than 30 million users. In November 2025, Luma closed a $900 million Series C round led by HUMAIN an AI company backed by Saudi Arabia’s Public Investment Fund pushing its valuation above $4 billion. Total funding now exceeds $1 billion, with investors including Andreessen Horowitz, Amazon, AMD Ventures, and NVIDIA.

In March 2026, Luma launched its AI Agents the tools now at the center of the Innovative Dreams workflow. These agents are designed for end-to-end creative production: they can handle text, image, video, and audio in an integrated pipeline, rather than requiring separate specialized tools for each task. The launch of Innovative Dreams is, in part, a live demonstration of what those agents can actually do at scale. This connects directly to the broader trend of AI agents moving from experimental features into full production workflows.

For Jain, the Hollywood pivot is an extension of a thesis he has been articulating publicly for a while: that soaring production costs have made filmmaking increasingly risk-averse, and that generative AI can break that logjam not by removing creativity from the equation, but by removing the financial ceiling that limits what creative teams can attempt.

What Innovative Dreams Could Mean for AI Video as a Category

The most significant thing about Innovative Dreams isn’t any single technical achievement. It’s the model it establishes. For years, the standard AI-in-Hollywood narrative has been a licensing story: AI companies license their models to studios, studios use them to reduce VFX costs, everyone makes money. That model works, but it keeps the AI company at arm’s length from the actual creative process.

Innovative Dreams is a different model. Luma is co-producing. Luma’s creative technologists are embedded alongside Erwin’s filmmakers. The company isn’t just providing infrastructure it’s participating in the result. That changes the feedback loop significantly. When Luma’s tools are used to make a film that Luma has a stake in, every technical decision becomes a creative decision. The incentive to actually solve hard production problems not just provide tools that sort-of-work is fundamentally different.

It’s also a smart way to generate training data for future model improvements. Every creative decision made during production of “The Old Stories: Moses” every lighting adjustment, every set modification, every performance capture session feeds information back into Luma’s understanding of what photorealistic AI-generated film content actually needs to look like. That’s a competitive advantage that can’t be replicated by a company that only licenses its tools and never touches the finished product. This is part of the same broader story we’ve been tracking the way real-time generative AI is transforming visual media, from gaming to now Hollywood production.

What Comes Next

“The Old Stories: Moses” will be the first real test of Luma’s real-time hybrid filmmaking claims. If Ben Kingsley’s performance lands and the visual scope holds up at the scale of a Biblical epic, Innovative Dreams will instantly become one of the most cited case studies in AI-assisted production. If the cracks show, the skeptics will have their evidence.

But either way, the move matters. The AI video generation space has spent the last two years building increasingly capable tools. Luma, Runway, Higgsfield, Wonder Studios they’re all now in the business of proving those tools can make something worth watching. That’s a harder challenge than shipping a model with impressive benchmark scores. And the companies that can demonstrate real creative results, with real Hollywood professionals and real audiences, are going to have a different conversation with the studios than the ones still selling promises on a slide deck.

The faith audience Innovative Dreams is initially targeting is, in some ways, the perfect proving ground. They show up. They engage. They spread word-of-mouth. If “The Old Stories: Moses” resonates with that community, Luma won’t just have a production company it’ll have a distribution moat that most AI startups would struggle to build even with ten times the funding. The AI-in-Hollywood story is getting more interesting by the week. And Luma just made sure it’s going to be telling a significant part of that story.