OpenAI is founded as a nonprofit AI research lab

Background

In the years before OpenAI was launched, work on neural networks had moved from a narrow academic topic to the centre of the field, and a small circle of researchers began arguing that the direction of the technology should not be left entirely to commercial incentives. OpenAI was created to try a different arrangement: a research laboratory set up as a non-profit, with its mandate framed around public benefit rather than financial return^[1].

The announcement

OpenAI was founded in December 2015 as the non-profit OpenAI, Inc., headquartered in San Francisco^[1]. Its founding announcement, dated 11 December 2015, described it as a non-profit artificial intelligence research company^[4].

The goal it stated was "to advance digital intelligence in the way that is most likely to benefit humanity as a whole, unconstrained by a need to generate financial return"^[1][4]. A shorter formulation of the mission is that it exists "to ensure that artificial general intelligence benefits all of humanity"^[1].

The founders

The founding group mixed entrepreneurs with research scientists. Elon Musk and Sam Altman were co-chairs; the other founders were Ilya Sutskever, Greg Brockman, Trevor Blackwell, Vicki Cheung, Andrej Karpathy, Durk Kingma, John Schulman, Pamela Vagata and Wojciech Zaremba^[1].

The announcement itself gave several of them formal roles: Sam Altman and Elon Musk as co-chairs, Ilya Sutskever as research director, and Greg Brockman, formerly chief technology officer of Stripe, as chief technology officer^[4].

Two of the founders' moves into the project are recorded. Sutskever left Google at the end of 2015 to co-found OpenAI as its chief scientist^[3]. Brockman had left Stripe in May 2015, and after co-founding OpenAI went on to lead projects including OpenAI Gym and OpenAI Five^[2].

Money

At launch the funders had committed $1 billion, although the announcement said OpenAI expected to spend only a tiny fraction of that in the next few years^[4]. The distance between the pledge and the cash was considerable: according to tax filings, only $133 million had been received by 2021^[1].

That gap is a useful reminder of how ambitious research organisations are financed in practice — a headline figure signals intent and attracts attention, while the money that actually arrives sets the real scale of the work.

Departures and continuity

Elon Musk left the OpenAI board in 2018^[1]. The laboratory's work continued through the people who stayed: Brockman, for instance, later led the OpenAI Gym and OpenAI Five projects^[2], and Sutskever had come in as chief scientist from Google at the end of 2015^[3].

Why it matters

OpenAI was founded with two features that were unusual together. It named artificial general intelligence — the long-term goal of machines that can do intellectual work across domains — as its target, and it wrote public benefit into its founding statement rather than return to investors^[1]. Putting that aim in a public announcement made the long-term question part of the wider conversation about AI, not only a subject for specialists.

It also showed that a non-profit could draw researchers out of established labs: Sutskever left Google and Brockman left Stripe to join at the start^[3][2]. The founding roster, with a dozen named people holding roles from co-chair to chief scientist, set the pattern of a small technical core surrounded by outside supporters.

Finally, the founding episode is a case study in how announcements and reality diverge. A billion dollars was committed and only a fraction of it had arrived years later^[4][1], while the organisation's own description of itself — non-profit, San Francisco, benefit to humanity — became the reference point against which its later changes would be measured^[1].