It's a theory, but it's not like they refrained from badmouthing him and kept it all professional. That would have looked like "after careful deliberation, we have determined that OpenAI cannot meet its mission under current management, and..."

It's extremely strange to make it clear the firing is for cause, and to be vague about the cause, and then to issue statements that whatever the cause was it was not related to fiduciary issues or AI safety. It really does amount to "he did something really bad that created a crisis we had no choice but to respond to precipitously... but it wasn't all that bad and we don't want to talk about it" situation.

I think "incompetent seems likely" is the only reasonable conclusion here.