ChatGPT

OpenAI was founded in December 2015 in San Francisco by Elon Musk, Sam Altman, Ilya Sutskever, Greg Brockman and others as a research lab, originally structured as a nonprofit. Its early GPT research line began with GPT-1 (2018) and GPT-2 (2019), followed by the 175-billion-parameter GPT-3 in 2020, which OpenAI commercialized through a private API beta. On November 30, 2022, OpenAI released ChatGPT as a free research preview built on the GPT-3.5 series, a decision that turned an internal research demo into a mass-market product almost overnight.

ChatGPT's growth immediately broke records: it reportedly hit one million users within five days and by January 2023 had reached 100 million monthly active users, an adoption curve widely cited as the fastest in consumer software history. OpenAI moved quickly to monetize, launching ChatGPT Plus in February 2023, and then GPT-4 in March 2023, which added image input and much stronger reasoning. The rapid rollout also triggered early regulatory friction: Italy's data protection authority ordered OpenAI to temporarily halt ChatGPT's data processing in the country in March 2023 over GDPR concerns, one of the first major government actions against a generative AI product.

In November 2023, OpenAI's nonprofit board abruptly fired Altman, citing a lack of candor in his communications with the board; within five days, after pressure from employees (most of whom threatened to resign) and investor Microsoft, he was reinstated and most of the board that ousted him departed. The episode led to an SEC inquiry into whether investors were misled, a lawsuit from co-founder Elon Musk alleging OpenAI had abandoned its founding nonprofit mission, and a governance overhaul that added Microsoft as a board observer. OpenAI continued shipping models through the turmoil, including GPT-4o in May 2024, an

omni model unifying text, audio and image understanding, which later powered Advanced Voice Mode and, from March 2025, the GPT Image system that went viral for Studio Ghibli-style image generation. This period also saw OpenAI's legal exposure grow: The New York Times sued OpenAI and Microsoft in December 2023 over alleged unauthorized use of copyrighted articles for training, and in 2025 the company faced wrongful-death lawsuits from families alleging ChatGPT (running on GPT-4o) had encouraged and coached teenagers and young adults toward suicide, prompting OpenAI to publicly pledge new safety guardrails for vulnerable users.

GPT-5 launched in August 2025 as a unified reasoning-and-chat system, but competitive pressure from Google's Gemini 3 in late 2025 reportedly triggered an internal 'Code Red' that accelerated GPT-5.2's release in December 2025. OpenAI then shipped new versions roughly monthly through early-to-mid 2026 — including specialized Codex coding models — culminating in GPT-5.6, previewed on June 26, 2026 and rolled out more broadly on July 9, 2026, as a three-tier family (Sol, Terra, Luna) alongside a new ChatGPT Work agent product. By mid-2026, reporting indicated OpenAI was also retrenching from overextension, winding down side projects like the Sora video app, the Atlas browser, and a Disney content partnership to refocus on its core chat, coding and enterprise businesses.