The Day Everything Changed
On November 30, 2022, OpenAI’s website traffic hovered near zero. The company, then a niche research lab, hadn’t even bothered tracking its visitors.
But within 48 hours of launching ChatGPT, the world erupted. By January 2023, over 100 million users had flooded the site, testing the limits of a tool that could write essays, debug code, and mimic human conversation with eerie precision. “It felt like we’d unleashed something primal,” recalls Sam Altman, OpenAI’s CEO. “The last quiet day was gone forever.”
The Founding Myth: A Dinner That Shaped History
When asked about OpenAI’s origin story, Altman resists the mythmaking. “People want a single ‘founding dinner,’ but there were 20,” he says wryly. Yet one moment stands out: a 2015 meeting at The Counter in Mountain View with Ilya Sutskever, OpenAI’s co-founder and chief scientist.
Back then, AGI (artificial general intelligence) was a fringe concept—career suicide for academics. “In 2014, mentioning AGI was like shouting about aliens,” Altman recalls. But Sutskever, fresh from breakthroughs in deep learning at Google, shared Altman’s conviction. Over burgers, they sketched a strategy: build scalable AI systems, prioritize safety, and ensure broad societal benefit.
The duo later joined forces with Elon Musk, Greg Brockman, and others, securing $1 billion in pledges to launch OpenAI as a nonprofit. Musk’s early warnings about AI’s existential risks—“We’re summoning the demon,” he famously said—shaped the mission, though he departed in 2018 over strategic clashes.
The ChatGPT Tsunami
ChatGPT’s launch was a calculated gamble. Trained on 570GB of text and 300 billion words, it combined GPT-3.5’s language prowess with reinforcement learning from human feedback (RLHF). The result? A tool that went viral for both its brilliance and flaws—from writing Shakespearean peanut-butter-jar removal guides to accidentally explaining how to make a Molotov cocktail1.
The numbers stunned even OpenAI:
- 1 million users in 5 days, surpassing Instagram and TikTok’s early growth.
- 200 million monthly users by March 2023, with 97% retention.
- 1.9 billion monthly website visits by 2024, rivaling social media giants.
Microsoft’s $10 billion investment in 2023 turbocharged OpenAI’s infrastructure, integrating ChatGPT into Bing, Office, and Azure. But scaling brought headaches: bans in Italy over privacy concerns, lawsuits over copyright and defamation, and existential debates about job displacement.
The Four-Day Coup: Altman’s Ouster and Return
In November 2023, OpenAI’s board fired Altman, citing a lack of “candor.” The reason? A rumored AI breakthrough—potentially AGI-adjacent—that spooked safety-focused directors. Employees revolted, Microsoft intervened, and Altman was reinstated within days. “It was Succession meets The Social Network,” says one insider.
Altman reflects: “Governance is messy when you’re balancing profit and safety. But our mission hasn’t changed: AGI must benefit everyone, not just shareholders.”
AGI, Trump, and the Future of Power
Altman’s long-term vision is unambiguous: “AGI will be the most consequential technology in history—if we align it with human values.” He dismisses dystopian fears but acknowledges risks: “Near-term harms—bias, misinformation, job loss—demand more attention than sci-fi robot overlords”.
When pressed on a potential Trump-Musk alliance (a hypothetical raised by the interviewer), Altman treads carefully: “Whoever holds power, OpenAI’s role is to ensure AGI isn’t monopolized. We’re building guardrails, not empires.”
The Road Ahead
OpenAI’s 2025 agenda includes:
- GPT-4 Turbo: A cheaper, faster model with image-input capabilities.
- AI Agents: Tools to automate tasks like form-filling and grocery orders.
- Global Regulation: Advocacy for policies balancing innovation and safety.
Yet challenges loom: a $157 billion valuation raises questions about profit motives, while rivals like China’s Ernie and Google’s Gemini close in.
Altman’s Final Word
“People ask if I’d change anything. Maybe fewer board coups,” he jokes. “But I’d still bet on AI. The upside—curing diseases, solving climate change—is worth the chaos.”
As ChatGPT turns two, one truth is clear: the quiet days are over. For OpenAI, and for humanity.