Skip to content

Sam Altman Regrets Nonprofit Status for OpenAI

In an otherwise boring hour-long “fireside chat” at Harvard about 6 months ago, OpenAI (c)(3) founder Sam Altman discussed the thinking process that went into the founding of OpenAI (c)(3) and the whole for-profit network referred to simply as “OpenAI.” Jill, Ellen, and Rose describe that network in recent commentary:

OpenAI began in 2015 as a scientific research nonprofit. Four years later, its board decided that achieving its lofty goals required more than gifts and grants.  It reorganized to accommodate and attract private investment. As a result, the company known as OpenAI is neither a single nonprofit nor for-profit company; it is a set of interlocking entities, including the for-profit subsidiary that conducts its operations.

At around the 15:00 mark, Altman pretty much admits that he and his cofounder Ilya Sutskever selected the original nonprofit model for lack of a product. Almost as a mere placeholder.  They didn’t know exactly what they wanted to build or discover, how much it might cost, and whether it would be profitable so they thought they would just conduct research and write papers until something came up.  After “stumbling around in the dark” for awhile — presumably with start-up charitable contributions — they discovered the infinite potential of “large language models” and suddenly realized they needed “gigantic amounts of funding.”

So they went looking for a business model.  From the brief conversation, it seems clear that they were all fascinated by the scientific possibilities, but hardly concerned about grander ideas of benefiting humanity.  That part is hardly ever mentioned.  The omission doesn’t impeach their altruistic motives, whatever they may be, as much as it demonstrates that the need for “gigantic amounts of funding” puts limits on charitable intentions.  In fact, at around 53:15, Altman wistfully admits that if he could go back in time, he would select a for-profit business model from the start.  I wonder if Elon was paying attention.  

darryll k. jones