While browsing YouTube, I stumbled across a video titled This Book Changed How I Think About AI. Curious, I clicked and it introduced me to Empire of AI by Karen Hao, a book that dives deep into the evolution of OpenAI.
The book explores OpenAI’s history, its culture of secrecy, and its almost single-minded pursuit of artificial general intelligence (AGI). Drawing on interviews with more than 260 people, along with correspondence and internal documents, Hao paints a revealing picture of the company.
After reading it, I uncovered 12 particularly fascinating facts about OpenAI that most people don’t know. Let’s dive in.
1. The “Open” in OpenAI Was More Branding Than Belief
The name sounds noble who doesn’t like the idea of “open” AI? But here’s the catch: from the very beginning, openness was more narrative than commitment. Founders Sam Altman, Greg Brockman, and Elon Musk leaned into it because it helped them stand out. Behind closed doors, though, cofounder Ilya Sutskever was already suggesting they could scale it back once the story had served its purpose. In other words: open, until it wasn’t convenient.
2. Elon Musk’s Billion-Dollar Promise? Mostly Smoke and Mirrors
Remember Musk’s flashy $1 billion funding pledge? Turns out, OpenAI only ever saw about $130 million of it. And less than $45 million came directly from Musk himself. His back-and-forth on funding almost pushed the organization into crisis, forcing Altman to hunt down new sources of money.
3. The For-Profit Shift Was More About Survival Than Vision
In 2019, OpenAI unveiled its “capped-profit” structure, pitching it as an innovative way to balance mission and money. But the truth is far less glamorous: the nonprofit model wasn’t bringing in the billions needed to compete with tech giants. At one point, Brockman and Sutskever even discussed merging with a chip startup. Creating OpenAI LP wasn’t a bold visionit was a lifeline.
4. The “Capped-Profit” Model Looked Unlimited to Critics
Investors were told their returns would be capped at 100x. Sounds responsible, right? But do the math: a $10 million check could still turn into a $1 billion payout. Critics quickly called it “basically unlimited,” arguing the cap only looked meaningful until you saw the actual numbers.
5. GPT-2’s “Too Dangerous” Storyline Was a PR Masterstroke
In 2019, OpenAI said its GPT-2 model was so powerful it had to be withheld for safety reasons. Headlines exploded. But here’s the twist: many researchers thought the risk claims were overblown and saw the whole thing as a publicity stunt engineered by Jack Clark, OpenAI’s communications chief at the time. The stunt worked—the company was suddenly everywhere.
6. OpenAI’s Culture Had Clashing “Tribes”
Inside OpenAI, things weren’t exactly harmonious. Sam Altman himself described the organization as divided into three factions: research explorers, safety advocates, and startup minded builders. He even warned of “tribal warfare” if they couldn’t pull together. That’s not just workplace tension it’s a sign of deep conflict over the company’s direction.
7. ChatGPT’s Global Debut Was Basically an Accident
Think ChatGPT’s launch was carefully choreographed? Not at all. The product that made OpenAI a household name was released in just two weeks as a “research preview,” right after Thanksgiving 2022. The rush was partly to get ahead of a rumored chatbot from Anthropic. Even Microsoft OpenAI’s biggest partner was caught off guard and reportedly annoyed.
8. Training Data Included Pirated Books and YouTube Videos
Where do you get enough data to train something like GPT-3 or GPT-4? In OpenAI’s case, by scraping almost everything it could. GPT-3 used a secret dataset nicknamed “Books2,” which reportedly included pirated works from Library Genesis. GPT-4 went even further, with employees transcribing YouTube videos and scooping up anything online without explicit “do not scrape” warnings.
9. “AI Safety” Initially Ignored Social Harms
OpenAI loves to talk about AI safety now. But early on, executives resisted calls to broaden the term to include real-world harms like discrimination and bias. When pressed, one leader bluntly said, “That’s not our role.” The message was clear: safety meant existential risks, not everyday impacts.
10. Scaling Up Came with Hidden Environmental Costs
Bigger models require more compute and more resources. Training GPT-4 in Microsoft’s Iowa data centers consumed roughly 11.5 million gallons of water in a single month, during a drought. Strikingly, Altman and other leaders reportedly never discussed these environmental costs in company-wide meetings.
11. “SummerSafe LP” Had a Dark Inspiration
Before OpenAI LP had its public name, it was secretly incorporated as “SummerSafe LP.” The reference? An episode of Rick and Morty where a car, tasked with keeping Summer safe, resorts to murder and torture. Internally, it was an ironic nod to how AI systems can twist well-meaning goals into dangerous outcomes.
12. Departing Employees Faced Equity Pressure
Leaked documents revealed OpenAI used a hardball tactic with departing employees: sign a strict nondisparagement agreement or risk losing vested equity. This essentially forced people into lifelong silence. Altman later said he didn’t know this was happening and was embarrassed, but records show he had signed paperwork granting the company those rights a year earlier.
Final Thoughts
OpenAI’s story is anything but straightforward. From broken promises and internal clashes to controversial data practices, the company has often operated in ways that don’t match its public messaging. Whether you see that as savvy strategy, messy growing pains, or something more troubling depends on your perspective.
But one thing’s clear: the “open” in OpenAI has always been complicated.
Top comments (0)