I think I have to disagree here because they talk about loss of trust. Loss of trust is enough when deciding who you want to work associate with. Firing seems steep, but when working with someone feels forever untenable, your options are really you or them exiting. There's not really an in-between option to balance the response for how much the issue crossed the line of trust when you don't see it coming back.
Maybe they shouldn't have lost total trust at Sam's behaviour. But personally I'd lean towards it being a pretty normal response from people existing in the environment itself; whom are at the time feeling the active shift effect Sam was having on the power balance to their detriment.
This is an important point. OpenAI is a nonprofit who's stated mission is to ensure that AI benefits all of humanity. Overseeing the business entity that exists under it's umbrella, and ensuring that the actions it takes are in accordance with that mission, is THE critical component of the boards job. This structure is very weird for a non-profit, and the stakes here are existential.
If you cannot trust the CEO of the company to not deceive and manipulate you, you absolutely cannot have confidence that the companies actions will conform to their mission.
Altman is a sociopath and they should have never caved to his political machinations
What's not clear is what special sauce he brings, beyond being a slippery snake, that someone else couldn't bring.