Peter Steinberger spent a weekend building something the world wanted. Within weeks, OpenClaw had 196,000 stars on GitHub and 2 million people using it every week. Now he is joining OpenAI, and what happens next tells you everything about where artificial intelligence is heading.
This is not just another acquisition story. This is about what happens when the most valuable AI company on the planet realizes that community-built tools are outpacing their billion-dollar research labs. This is about whether open source can survive in an industry burning through money faster than any startup in history.
The Weekend Project That Changed Everything
OpenClaw started as Clawdbot, then became Moltbot, before settling on its final name. Steinberger, an Austrian software developer, launched it in November 2025. The concept was simple but powerful. While ChatGPT could write you a poem or explain quantum physics, it could not actually do things for you. It could not book your flight, clean your inbox, or manage your calendar.
OpenClaw changed that equation entirely. It was an AI agent that could take control, make decisions, and execute tasks without constant human oversight. You could tell it to organize your Spotify playlists, and it would. You could ask it to find and purchase the best deal on running shoes, and it would handle the entire process.
The response was immediate and overwhelming. Developers grabbed it, modified it, and built custom agents for everything imaginable. Chinese users paired it with DeepSeek language models. Businesses integrated it into their workflows. Baidu announced plans to add OpenClaw directly into their smartphone app.
The Acqui-Hire That Was Not Really an Acquisition
Sam Altman posted on X that Steinberger would be joining OpenAI to work on personal agents. He called him a genius with amazing ideas about smart agents working together. The deal was reportedly worth between 2 and 15 million dollars, though neither party confirmed exact figures.
But here is where it gets interesting. This was not OpenAI buying OpenClaw the product. They were buying OpenClaw the phenomenon, the community, the momentum. Steinberger himself was refreshingly honest about his motivations. He wrote on his blog that he could see OpenClaw becoming a huge company, but that prospect did not excite him.
“What I want is to change the world, not build a large company,” he explained. Thirteen years running his previous company had taught him what he did and did not want. OpenAI offered him the resources to make agents accessible to everyone without the burden of building another corporate empire.
Meta had also made an offer, reportedly in the billions. Multiple major labs courted him during a week he spent in San Francisco. The decision came down to which company would let OpenClaw remain truly open source. Steinberger chose OpenAI because they promised the project would live in an independent foundation with continued support.
What Open Source Really Means in 2026
OpenAI announced that OpenClaw will operate through a foundation as an open source project. Altman emphasized that the future would be extremely multi-agent and that supporting open source was crucial. These are the right words, carefully chosen.
But words are one thing. Reality is another. OpenAI began life in December 2015 as a nonprofit research lab. The founding team included Elon Musk, Sam Altman, Greg Brockman, and Ilya Sutskever. They raised over a billion dollars from luminaries like Peter Thiel and organizations like Amazon Web Services.
The stated mission was clear and idealistic. They would develop artificial general intelligence that benefited all of humanity. They would collaborate openly with other institutions. They would make their patents and research public. OpenAI was supposed to be different from the profit-hungry tech giants.
That lasted exactly four years. In 2019, OpenAI created a capped-profit subsidiary. The reasoning was pragmatic. Advanced AI research required massive computing power and sustained funding that nonprofit structures could not support. Top researchers were going to Google Brain and DeepMind and Facebook, companies that offered equity stakes a nonprofit could never match.
The capped-profit model allowed investors to earn returns up to 100 times their investment. Everything beyond that would flow back to the nonprofit. Microsoft invested a billion dollars and became the primary partner, providing cloud infrastructure and computing credits. OpenAI systems now run exclusively on Azure.
From Nonprofit to Five Hundred Billion Dollar Valuation
By October 2025, OpenAI completed another restructuring. The for-profit arm became OpenAI Group PBC, a public benefit corporation. The nonprofit foundation maintained control and held a 26 percent equity stake worth approximately 130 billion dollars.
The company is now valued at 500 billion dollars. ChatGPT has over 800 million weekly users. OpenAI is one of the fastest-growing commercial entities on the planet. Everything changed when they released ChatGPT in November 2022.
But growth and value are not the same as profit. OpenAI is hemorrhaging money at a scale that defies comprehension. In 2024, the company generated somewhere between 3.7 and 4 billion dollars in revenue. It lost 5 billion dollars. That means they spent 9 billion dollars to make 4 billion dollars.
The math gets worse in 2025. Revenue projections range from 13 to 20 billion dollars. The company expects to lose between 9 and 14.4 billion dollars. They are spending approximately 1.69 to 2.25 dollars for every dollar they earn. No startup in history has operated with losses on this scale.
Where Does All The Money Go
The costs are staggering and specific. Training new AI models cost OpenAI roughly 3 billion dollars in 2024. Running those models for users, the inference costs, added another 2 billion dollars. The compute budget alone exceeds their entire revenue.
Research and development consumed 6.7 billion dollars in just the first half of 2025. Salaries topped 700 million dollars before considering stock-based compensation, which added another 2.5 billion dollars. Sales and marketing ate up 2 billion dollars in six months.
Microsoft disclosures revealed that OpenAI lost 12 billion dollars in the quarter ending September 2025. That is a single quarter. The company is burning through approximately 575,000 dollars every hour.
OpenAI projects that cumulative losses will reach 143 billion dollars by 2029. They expect operating losses to hit 74 billion dollars in 2028 alone. The company says it will become profitable in 2029 or 2030, reaching 200 billion dollars in annual revenue.
The Economics Make No Sense Until They Do
Critics point out that OpenAI loses money on every single customer, both free and paid. They have 15.5 million paying subscribers, but acquiring and serving those customers costs more than the subscription fees generate. Adding more paying customers somehow increases the burn rate rather than decreasing it.
The company does not have diversified revenue streams like Google or Meta or Amazon. Those companies can lose billions on experimental projects while their advertising or cloud businesses fund the losses. OpenAI has subscriptions and API access. That is it.
Yet investors keep writing checks. In April 2025, OpenAI raised 40 billion dollars at a 300 billion dollar valuation, the largest private funding round in history. In January 2026, the company announced The Stargate Project, a joint venture with Oracle and SoftBank to build AI infrastructure systems worth 500 billion dollars.
Why would rational investors pour money into a company burning cash this fast? Because they believe AGI is coming.They believe OpenAI will achieve artificial general intelligence, that mythical system that matches or exceeds human cognitive abilities across all domains. They believe whoever gets there first will control the future.
Sam Altman has said that curing cancer with AI would require 7 trillion dollars and 10 gigawatts of power. He has sketched out plans for infrastructure spending that exceeds a trillion dollars. The scale is incomprehensible, but so are the potential returns if AGI becomes real.
The Competition Is Not Waiting
OpenAI faces intense competition from every direction. Anthropic, founded by former OpenAI researchers Dario and Daniela Amodei, raised funding in February 2026 that valued the company at 380 billion dollars. Their Claude models are gaining enterprise traction, particularly with the Claude Code tool that lets AI agents write and execute code.
Google has Gemini. Meta has Llama and an entire open source ecosystem. Elon Musk, who left OpenAI in 2018 after conflicts over direction, founded xAI and is raising 15 billion dollars at a 230 billion dollar valuation.
The difference is that competitors like Anthropic are spending more conservatively. Anthropic expects to break even by 2028 while OpenAI projects mounting losses through that same period. Anthropic is avoiding expensive ventures like image and video generation that require massive computing resources. Their costs grow roughly in line with revenue.
OpenAI is doing the opposite. They launched Sora 2 for video creation, which reportedly costs millions of dollars per day to operate. They are building a web browser called Atlas. They are developing consumer hardware with Jony Ive. They are researching humanoid robots. They are adding e-commerce and advertising features to ChatGPT.
This is either visionary diversification or reckless spending. We will know which within 24 months.
What This Means For You
If you use ChatGPT, expect price increases. The current subscription fees do not cover costs. OpenAI will either raise prices significantly or find entirely new business models. Enterprise customers with deep pockets will subsidize consumer access, or consumer access will become limited.
If you build on OpenAI APIs, expect volatility. The company needs revenue growth that matches spending growth. API pricing will fluctuate. Terms will change. OpenAI might prioritize certain use cases over others based on profitability rather than innovation.
If you care about open source AI, watch what happens to OpenClaw. Altman promised it would remain open and independent in a foundation. Those are the same kinds of promises OpenAI made about itself in 2015. The company had to make compromises to survive and scale.
Will OpenClaw face similar pressures? Will the foundation have genuine independence or will it gradually become another OpenAI product? The project can technically remain open source while OpenAI captures all the commercial value through integration and hosting and enterprise features.
The Deeper Question Nobody Wants to Answer
Here is what keeps some people awake at night. What if the economics never work? What if running advanced AI systems at scale is fundamentally unprofitable at any price point users will pay?
The entire AI industry is built on a bet that demand will eventually catch up to costs. Companies are spending 1.69 dollars to earn 1 dollar today because they believe they will spend 90 cents to earn 1 dollar tomorrow. They are building infrastructure for a future where everyone needs AI for everything.
But what if that future is slower to arrive than expected? What if AI capabilities plateau before they justify the investment? What if users love free AI but refuse to pay premium prices?
Deutsche Bank analysts noted that OpenAI is in completely uncharted territory. Uber lost 18 billion dollars over six years before profitability. Amazon took five years and a billion dollars. Tesla needed nine years and 9 billion dollars. OpenAI is projecting losses that dwarf all of those combined.
The only comparable might be the dot-com bubble, when telecom companies extended financing to customers to buy their own equipment. Money flowed in circles. NVIDIA invests in OpenAI. OpenAI signs massive cloud contracts with Oracle. Oracle buys chips from NVIDIA. The same dollars chase themselves around the industry.
What Actually Matters Right Now
Steinberger built something people wanted using open source tools and community collaboration. He did not need billions in funding or thousands of employees or exclusive partnerships. He spent a weekend solving a real problem in a way that made sense.
OpenAI recognized that community-driven innovation was moving faster than their internal research in certain areas. They made the smart move and brought Steinberger into the fold. Whether that was the right move for the open source community is a different question entirely.
The next few years will reveal whether the AI industry is building something sustainable or burning through the largest pile of capital ever assembled on a bet that might not pay off. The answer matters for everyone, not just investors and founders.
OpenClaw will either flourish as a truly independent open source project, or it will become another case study in how promising technologies get absorbed into the commercial machine. We have seen this story before. The ending is rarely the one that was promised.
Right now, you can still download OpenClaw, run it on your own hardware, modify it however you want, and build your own agents. Enjoy that while it lasts. Technology has a way of moving from “free and open” to “free tier with premium features” to “contact sales for enterprise pricing.”
The question is not whether OpenAI will try to monetize OpenClaw. They will. The question is whether the open source community and the foundation structure will have enough independence to push back when that inevitable moment arrives.
Time will tell if OpenAI remembers what “open” means, or if that word has already become just another piece of marketing wrapped around a very expensive, very closed, very traditional technology company racing toward a profitability target it might never actually reach.