Anyway, my question is - Have you had a chance to research OpenAI and it's path to getting "corrupted" i.e. going from Non-profit to this weird structure to then, what it looks like, absolutely regular for-profit company? whats your take on it? what are the lessons of the convoluted company structure they persuaded?
It's a really complicated and confusing company to draw lessons from, partly because there are such big ego, larger-than-life people involved doing crazy stuff. A lot of the inner details have not yet been made public, so I don't think we really, really know what was going on behind the scenes. There have been quite a few factional struggles there, and I'm sure you can find bad behavior on all sides. That's not that uncommon when so much money is at stake.
In the book, I try to get into the structure that seems to be stable over long periods of time, which, because I know more about the details, I felt more comfortable illustrating by telling the story of Anthropic. I had to pick one difference that seems to really matter in these so-called complicated structures. It's whether there is one central point of control or more than one, i.e., if there are checks and balances. That when an outside board of trustees has the power to hold the for-profit board accountable to the mission, that seems to produce more stable results.
Check out the research of Steve Thompson at the Copenhagen Business School, who has been a pioneer in developing the data from which these conclusions are drawn.
Thanks for the pointer. I’ll check it out.
Excited to read the book! Ordering mine today…