The surprise is more at the (EDIT: brazen) pathological lying.
This is only the 2nd or 3rd thing that seems to me even a little incoherent with their initially stated position, the other certain one being the mystery of why the board couldn't find anyone to replace Altman who didn't very quickly decide to take his side, and the other possible one being asking for a profit making subsidiary to raise capital (though at the time all the criticism I remember was people saying they couldn't realistically reach x100 and now it's people ignoring that it's limited to only x100).
They state something publicly, but are headed to completely different trajectory in reality.
This is enough for me.
In fact, I would consider changing goals publicly to be better than not following the goals.
Wired claims OpenAI’s “reports to US tax authorities have from its founding said that any member of the public can review copies of its governing documents, financial statements, and conflict of interest rules.” That was apparently a lie.
> Changing their goals is also not lying
Changing a forward-looking commitment is. Particularly when it changes the moment it’s called.
I don't know if I'd put this in terms of a "lie" or not, but OpenAI's stated principles and goals are not backed up by their actions. They have mined other people's works in order to build something that they purport as being for the benefit of mankind in some way, when their actions actually indicate that they've mined other people's work in order to build something for the purpose of massively increasing their own power and wealth.
I'd have more respect for them if they were at least honest about their intentions.
That's only under the assumption that the split with Altman was due to the doomers vs bloomers conflict and not just a dirty move from OpenAI board member Adam D'Angelo, trying to protect his investment in Quora's AI Poe.
[1] RIP "they switched everyone to prepaid billing!!!11!" I ate probably -10 saying "no, you got that email saying it was available because it was a feature announced at devday as coming soon"
Except that's not really true. Almost everyone on the board were either VCs themselves or had very strong ties to the them. In any case OpenAI would be irrelevant without significant investments from organizations/people who want a return on them. So it's basically a moot point: no VCs/big corporations = no fancy,extremely expensive to train & develop LLMs.