In November 2025, Jeff Bezos stepped back into a hands-on CEO role for the first time since leaving Amazon’s day-to-day leadership in 2021. He co-founded Project Prometheus (commonly called Prometheus), an ambitious artificial intelligence startup focused on “physical AI” – applying advanced machine learning to the engineering and manufacturing of complex physical systems, including computers, automobiles, and spacecraft.
Just seven months later, in June 2026, the company announced a massive $12 billion Series B funding round at a $41 billion valuation, bringing total capital raised to roughly $18.2 billion. The round was led by major institutions including JPMorgan Chase, Goldman Sachs, BlackRock, DST Global, and Arch Venture Partners, with Bezos participating personally.
Prometheus has quickly become one of the best-funded AI startups in the world, drawing attention not only for its capital but for its singular mission: dramatically compressing the invention and engineering cycles that underpin technological progress and, by extension, societal wealth.
The Vision: An Artificial General Engineer
Bezos and co-founder/co-CEO Vik Bajaj describe Prometheus as building tools that act as a modern, AI-powered evolution of computer-aided design (CAD) systems – tools capable of accelerating the full “invention loop” from idea to manufacturable product.
In a June 2026 interview, Bezos framed the goal in sweeping historical terms:
The company is explicitly targeting an “artificial general engineer” – AI systems that can help (or in some cases autonomously handle) the complex, multi-disciplinary work traditionally requiring “a thousand human minds creatively working together.” Early focus areas include jet engine design, spacecraft engineering, automotive systems, advanced computing hardware, and even drug design.
Unlike many AI efforts centered on chatbots, content generation, or consumer applications, Prometheus is squarely aimed at the physical economy – the hard engineering and manufacturing domains that have seen relatively slow productivity gains compared with software. Bezos has emphasized that the company is not primarily building robots or factory-floor automation, but rather upstream tools that make engineers dramatically more effective at designing and iterating on complex physical products.
Development has been underway since late 2024. As of mid-2026, the company employs approximately 150 people across offices in San Francisco, London, and Zurich, with talent drawn from OpenAI, Google DeepMind, Nvidia, and other leading AI and engineering organizations.
Massive Backing And Valuation
The scale of capital is extraordinary for a company of its age and stage. The initial $6.2 billion round (largely backed by Bezos) was already one of the largest early-stage AI raises ever. The follow-on $12 billion Series B at a $41 billion post-money valuation underscores investor conviction in both the technical vision and the team’s ability to execute.
A significant portion of the capital is earmarked for compute resources. In the June 2026 CNBC interview, Bezos noted that the work is “very compute-intensive” because the team must generate the specialized data needed to train models capable of reasoning about physical systems, materials, manufacturing constraints, and multi-physics interactions. He has indicated that Prometheus sources compute from multiple providers (including AWS) and expects to be a major customer of hyperscalers as it scales.
Leadership Duo: Bezos And Bajaj
Bezos serves as co-CEO and has stated that Prometheus represents the bulk of his professional time, though he continues significant involvement with Blue Origin and AI initiatives at Amazon. He has been clear that Prometheus operates independently. In reporting on the company’s structure, Bezos noted that it “deserves a dedicated team that is obsessed with this one thing.”
His partner is Vik Bajaj, a physicist and chemist with deep experience at the intersection of AI, large-scale computation, and scientific discovery. Bajaj previously co-founded and served as Chief Scientific Officer of Verily (Google Life Sciences/Alphabet), was CSO at GRAIL (the early cancer detection company acquired by Illumina), and founded Foresite Labs, an AI and data-science incubator focused on life sciences. He brings expertise in applying advanced computation and AI to complex real-world scientific and engineering problems – precisely the bridge Prometheus needs between frontier AI models and hard physical systems.
Relevance To The Space Economy
For the space industry, Prometheus’ ambitions are particularly salient. Spacecraft design and manufacturing sit at the extreme end of engineering complexity – long development cycles, stringent reliability requirements, exotic materials, and tight integration across disciplines. Early reports explicitly listed spacecraft among the target domains alongside automobiles and computers.
Accelerating spacecraft and propulsion system design could meaningfully compress timelines for launch vehicles, satellites, habitats, and in-space infrastructure. In a sector where development cycles for major systems often span 5–10+ years, even modest AI-driven speedups could shift competitive dynamics, lower costs, and enable faster iteration – key enablers for scaling the broader space economy.
Blue Origin itself is cited by Bezos as a natural potential user of the technology. While Prometheus remains structurally separate, the overlap in engineering challenges (rockets, vehicles, complex manufacturing) creates obvious synergy potential without formal corporate integration.
More broadly, Prometheus’ focus on the physical economy aligns with foundational needs of a maturing space sector: advanced manufacturing techniques, supply-chain optimization, and rapid prototyping of hardware that must survive extreme environments. If successful, it could contribute to the horizontal “capabilities” layer (manufacturing, engineering tools, production systems) that underpins vertical space applications.
Outlook, Optimism, And Open Questions
Bezos has been notably optimistic about AI’s macroeconomic impact, arguing it will drive significant productivity gains, raise living standards, and ultimately allow many households greater choice in work-life balance rather than triggering mass unemployment. He advocates for “reasonable” regulation focused on applications rather than blanket restrictions that could stifle innovation.
At the same time, Prometheus remains early-stage. In the June 2026 CNBC interview, Bezos described recent progress as “really quite remarkable” but noted it is still “premature” to disclose specific accomplishments or timelines for product releases. The company has operated with a relatively low public profile until the recent funding announcement and joint interview with Bajaj. Bezos framed the decision to speak publicly as a response to natural curiosity after raising such large sums: “We’re not being secretive… We’re just being heads down and trying to do the work.”
Skeptics will rightly ask whether even well-funded AI can truly compress the messy, physics-constrained realities of engineering complex hardware as dramatically as software has been compressed. The gap between impressive demonstrations and production-grade, certifiable systems (particularly in aerospace) remains substantial. Prometheus will need to demonstrate not just better design tools, but tools that integrate with real-world manufacturing, testing, certification, and supply chains.
A Modern Prometheus
The name is deliberate. In Greek mythology, Prometheus brought fire – and with it knowledge and civilization – to humanity. Bezos and Bajaj are positioning their company as a catalyst for a new wave of invention, one in which AI becomes a powerful collaborator (or in some domains, a near-autonomous participant) in the creation of physical technologies that improve human life.
Whether Prometheus ultimately delivers on that civilizational framing remains to be seen. What is already clear is that one of the world’s most successful entrepreneurs has placed an enormous bet – financially, personally, and intellectually – on AI finally cracking the code of accelerated physical engineering. For the space industry and the wider industrial base, the implications could be significant.
Key Facts At A Glance
- Founded: November 2025 (work underway since late 2024)
- Co-CEOs: Jeff Bezos and Vik Bajaj
- Headquarters: San Francisco (offices in London and Zurich)
- Employees: Approximately 150 (as of June 2026)
- Total Funding: Roughly $18.2 billion
- Valuation (June 2026): Approximately $41 billion
- Core Focus: Physical AI / AI tools for engineering design and manufacturing of complex systems (including spacecraft)
- Notable Investors: JPMorgan Chase, Goldman Sachs, BlackRock, DST Global, Arch Venture Partners, and Jeff Bezos
The coming years will reveal whether Prometheus can turn its massive resources and ambitious vision into tools that reshape how humanity designs and builds the machines of the future – including those that will carry us deeper into space.
