💡 TL;DR - The 30 Seconds Version
💰 Former OpenAI CTO Mira Murati raised $2 billion for her six-month-old AI startup, breaking all seed funding records by over 4x.
🧠 Only 1,000 people worldwide can build cutting-edge AI models, creating a talent war where Meta offers $100 million signing bonuses.
🏃♀️ About 75 former OpenAI employees have started 30 companies that collectively raised $8 billion, creating the "OpenAI Mafia."
📈 AI companies raised $73.1 billion in Q1 2025, but deal volume fell to five-year lows as investors bet big on fewer teams.
🎯 Thinking Machines Lab has no product yet but commands a $10 billion valuation based purely on its team's OpenAI track record.
🚀 The funding shows top AI talent can now raise billions before building anything, reshaping how we value early-stage tech companies.
Mira Murati just broke Silicon Valley. The former OpenAI chief technology officer raised $2 billion for her six-month-old AI startup, setting a new record that makes every other seed round look tiny.
Thinking Machines Lab is now worth $10 billion. The company has no product, no revenue, and barely exists. But it has something more valuable: the people who built the AI revolution.
Murati didn't just leave OpenAI alone. She took some of the industry's best minds with her. John Schulman, who co-founded OpenAI, joined as chief scientist. Barrett Zoph, OpenAI's former research chief, became CTO. The team includes top researchers from Google, Meta, and Mistral AI.
This is the "OpenAI Mafia" in full swing. About 75 former OpenAI employees have started roughly 30 companies, raising almost $8 billion total. Murati's round alone accounts for a quarter of that money.
The previous seed funding record was $450 million. Murati's round is more than four times larger.
The Great AI Talent War
The AI industry has a simple problem: there aren't enough smart people to go around. Fewer than 1,000 people worldwide can build cutting-edge AI models. This scarcity has created a bidding war that makes professional sports look cheap.
Companies are throwing money at researchers like never before. Meta reportedly offers $100 million signing bonuses. Microsoft paid $650 million to buy Inflection AI mainly for its staff. Entire startups get acquired just for their teams.
Mark Zuckerberg is personally emailing hundreds of researchers. He's offering deals worth hundreds of millions to join Meta's new AI lab. Some candidates are saying no because they're worried about Meta's recent AI struggles and unclear leadership.
OpenAI CEO Sam Altman accused Meta of viewing his company as their biggest threat. The accusation came after Meta failed to buy several AI startups, including Thinking Machines Lab. Meta then changed tactics to focus on stealing executives instead of buying companies.
The talent war is getting expensive fast. Companies that can't attract top researchers are getting left behind. Those that can are paying unprecedented amounts.
When Venture Capital Goes Crazy
Murati's $2 billion round shows how much AI investing has changed. Venture capitalists are making fewer bets but much bigger ones. They're putting massive amounts of money on proven teams rather than spreading it around.
AI companies raised $73.1 billion in the first quarter of 2025. But the number of deals fell to a five-year low. Investors learned that most AI startups fail, so they're concentrating money on the few with real talent.
This creates a winner-take-all dynamic. Teams with strong track records can raise billions before building anything. Everyone else struggles to get attention.
Ilya Sutskever, OpenAI's former chief scientist, raised $1 billion for his new company at a $32 billion valuation. No product, no customers, just a great reputation. These mega-rounds are becoming normal for ex-OpenAI executives.
The numbers would have seemed impossible just a few years ago. Now they're just Tuesday in Silicon Valley.
The Andreessen Horowitz Bet
Andreessen Horowitz led Murati's round with help from Accel and Conviction Partners. The firm is betting that the people who built ChatGPT can do it again somewhere else.
This isn't just about money. It's about reshaping the AI industry. Every major tech company is racing to build better AI models. The competition is fierce, and the stakes keep rising.
Murati's team has a track record. They helped create the models that made OpenAI famous. Now they're free to build whatever they want with essentially unlimited funding.
The venture capital firm is betting that talent matters more than anything else in AI. If you have the right people, you can build the technology. If you don't, all the money in the world won't help.
A Crowded and Competitive Market
Thinking Machines Lab enters a packed field. Every major tech company has an AI division. Dozens of well-funded startups are chasing the same opportunities. The performance gaps between top models are shrinking.
Different companies have chosen different strategies. Anthropic focuses on safety for enterprise customers. Mistral AI champions open-source models. Google emphasizes integration with its existing products. OpenAI dominates consumer AI through ChatGPT.
Murati's company plans to focus on human-AI collaboration and scientific discovery. The team wants to build models that help with research and programming while staying safe and reliable.
The company's mission sounds similar to many competitors. The difference is execution. Murati's team has done this before. They know how to build models that actually work.
Murati also structured the company to give herself final say on all major decisions. She watched other AI companies struggle with investor pressure and mission drift. She's determined to avoid those problems.
The New AI Power Structure
The $2 billion round puts Thinking Machines Lab in elite company. Only a handful of AI companies are worth more than $10 billion. Most of them are backed by major tech companies or have been around for years.
Murati's startup joins that group before releasing its first product. This shows how much the AI landscape has changed. Success now depends more on people than products. The best researchers can raise billions based on reputation alone.
The OpenAI exodus is creating multiple well-funded competitors. These aren't just startups – they're potential new centers of power in the AI industry. They could challenge the dominance of Google, Microsoft, and other tech giants.
What This Means for AI
The massive funding round reflects a simple truth: AI is eating the world, and everyone knows it. Companies are willing to pay any price for the people who can build the future.
This creates both opportunities and risks. More competition should lead to better AI products. But the concentration of talent and money could also create new monopolies.
The race is no longer just about building better models. It's about controlling the people who can build them. Murati's round shows that the best talent can write their own ticket.
The Bigger Picture
Thinking Machines Lab represents more than just another well-funded startup. It's proof that the AI revolution is far from over. The industry is still young enough for new companies to challenge established players.
The $2 billion round also shows the limits of corporate control. OpenAI trained these researchers, but it couldn't keep them. Now they're building competing products with essentially unlimited resources.
This pattern will likely continue. The most talented AI researchers want to build their own companies rather than work for someone else. They have the skills, the connections, and now the funding to make it happen.
The question isn't whether more OpenAI alumni will start companies. It's how many, how fast, and how much money they'll raise. Based on Murati's success, the answers are: many, quickly, and lots.
Why this matters:
- The $2 billion round shows that top AI talent can command unprecedented funding, changing how we value early-stage technology companies and proving that people matter more than products in the AI race.
- The exodus of key OpenAI employees is creating multiple well-funded competitors that could break up the current AI oligopoly, giving us a glimpse of an industry where talent flows freely between companies rather than being locked up by tech giants.
❓ Frequently Asked Questions
Q: How does a $2 billion seed round compare to what most startups raise?
A: Most seed rounds are under $5 million. The previous record was $450 million by Yuga Labs in 2022. Murati's $2 billion round is larger than what many companies raise in their entire lifetime across all funding stages.
Q: Why are only 1,000 people worldwide capable of building cutting-edge AI models?
A: Building frontier AI models requires deep expertise in machine learning, massive computing infrastructure, and hands-on experience with large-scale training. Most researchers work on applications, not the foundational models themselves. The field is extremely new and specialized.
Q: What exactly is the "OpenAI Mafia" and how big is it?
A: About 75 former OpenAI employees have left to start roughly 30 new AI companies, collectively raising almost $8 billion. Major examples include Anthropic (founded by Dario Amodei), Safe Superintelligence (Ilya Sutskever), and now Thinking Machines Lab (Mira Murati).
Q: Has Thinking Machines Lab released any products yet?
A: No. The company is only six months old and has no products or revenue. It raised $2 billion purely based on its team's track record at OpenAI and their stated mission to focus on human-AI collaboration and scientific discovery.
Q: How much control does Mira Murati have over her new company?
A: Murati structured Thinking Machines Lab to give herself the deciding vote on all board matters. This level of founder control is unusual for such a large funding round, but investors agreed because of her proven track record.
Q: Are AI companies really offering $100 million signing bonuses?
A: According to OpenAI CEO Sam Altman, Meta has offered signing bonuses as high as $100 million to poach top researchers. Microsoft paid $650 million to buy Inflection AI mainly for its staff. The talent war has reached unprecedented levels.
Q: Why did venture capital invest so heavily in AI this quarter?
A: AI companies raised $73.1 billion in Q1 2025, but the number of deals fell to a five-year low. Investors learned most AI startups fail, so they're making fewer but much larger bets on proven teams with track records.
Q: When will we see products from Thinking Machines Lab?
A: The company hasn't announced a timeline. Based on typical AI development cycles and the complexity of their stated goals around scientific discovery and human-AI collaboration, products likely won't appear for at least 12-18 months.