OpenAI: the company that sparked the AI revolution
From nonprofit research lab to the most influential AI company in the world — OpenAI, ChatGPT, and the decade that changed everything.
No company has done more to shape public perception of AI than OpenAI — not by being first or always the most capable, but by making extraordinary technical work accessible. When ChatGPT launched on November 30, 2022, it became the fastest consumer product in history to reach 100 million users, in roughly two months. The capability had existed in the API for two years; what ChatGPT removed was friction.
The structure is the other half of the story. Founded in December 2015 as a nonprofit — Altman, Musk, Sutskever, Brockman, Zaremba, Schulman, with $1B pledged — its explicit worry was AGI as a dangerous concentration of corporate power. By 2019 the economics of frontier training forced a "capped profit" subsidiary, which unlocked Microsoft's eventual $13B+ and Azure's compute. In 2025 it restructured again into a public benefit corporation, under scrutiny from regulators, former employees, and Musk's lawsuit. The departures aren't noise — Sutskever leaving to found a safety-focused lab is part of an unresolved argument about whether the founding mission survived the funding model.
Technically, the through-line is the GPT lineage: GPT-2 was initially withheld over misuse concerns, GPT-3 (2020) reset the industry's expectations at 175B parameters, GPT-4 (2023) held the capability benchmark for over a year, and the GPT-5 family continues it. The o-series reasoning models are the notable architectural fork — generating extended internal reasoning before answering, trading cost and speed for accuracy on math, code, and analysis. Around the text models sit DALL·E 3, Sora for video, the widely deployed open-source Whisper, and one of the most mature developer platforms in the industry — for many teams still the default starting point.
Our read: OpenAI proved, in product terms, that capable AI can be useful to ordinary people, and that demonstration redirected an industry. Its release pace also represents a different risk tolerance than labs like Anthropic — a trade-off the industry hasn't settled. The hardest part of its story is the current one: keeping both capability leadership and the public trust that leadership requires.