That's pretty unsurprising for him to demand that, since the statement they put out really threw him under the bus. I would've thought that if he'd done something actually bad they would've come out and said it already, so as to prevent this whole power struggle.
Why on earth would they do that? That would just open them up to liability.
You're too used to this tech-sector 'put everything on twitter rn' bullshit immature behaviour.
Do you see the boards of GM, IBM, etc slapfighting on twitter? No, you don't, because it's abnormal.
Totally normal for a board to release a press release and then STFU. Weird twitter slapfights between CEOs and boards are not what we want to encourage. This isn't the pro-wrestling circuit.
The existing statements would be libel if false. It's a little surprising they put something like that in a press release at all; it would be shockingly inept if they did so without the sort of documentation that would make a lawsuit futile.
Of course, tho whole situation is a little shocking so maybe all bets are off.
Technically the only sentence in the statement that might be false is
"Mr. Altman’s departure follows a deliberative review process by the board, which concluded that he was not consistently candid in his communications with the board, hindering its ability to exercise its responsibilities."
And presumably this is why they used the "not consistently candid" weasel-wording. For that to be true, they only need to point to one material action that he didn't inform the board about (perhaps his chip startup, or Developer Day, or API access), and then point to the board meeting minutes, and then say "See, it's true. You did not inform us, the board, about these other actions you have taken, they matter to OpenAI's adherence to its mission, and therefore your inaction hindered our ability to exercise our responsibilities."
Everything else in the statement appears to be completely factual. It's a bad statement because it unnecessarily burns bridges and implies potential wrongdoing (because every other executive departure statement is so tame), but it didn't, strictly speaking, say anything untrue.
On the other hand, these words are widely understood to mean "he lied" and judges (or juries) aren't stupid. Going for this kind of strong wording and then trying to weasel out by a minor technicality seems like a risky strategy, especially against someone having sufficient money to fight this until the bitter end.
I hope we do. If these words would be on your recommendation letter or said by one of your job references, basically any HR person would understand it as "this employee was caught severely lying" and see it as a big warning not to hire you under any circumstances.
I agree that in a casual context, this would be mild, but we're talking about professional press releases and judges acting oblivious to this context is not a good thing in my book.
> The existing statements would be libel if false.
Libel isn't really an issue here, because the statement from the board hardly made any factual assertions in the first place:
> Mr. Altman’s departure follows a deliberative review process by the board, which concluded that he was not consistently candid in his communications with the board, hindering its ability to exercise its responsibilities. The board no longer has confidence in his ability to continue leading OpenAI.
There are a ton of possible interpretations of "he was not consistently candid".
However, there are lots of things that people can sue for besides outright libel/slander. I made the argument elsewhere that the legal nitpicking here isn't really that significant, because it's pretty easy to argue that the tenor of the board announcement would at least make it sound to a reasonable observer that Altman did something egregious.
The board said "he was not consistently candid in his communications with the board". That could mean little more than he once neglected to inform the board of something the board felt was important, which is almost certainly true. It's almost certainly true for any CEO; if you take it literally and don't read between the lines it's a very mellow accusation.
On HN and twitter people read that line and interpreted it to mean "he's a liar" or even "he raped his sister", but the board didn't actually say those things. He could try suing the board but I think he'd lose.
I haven't read them, but if they're accurate then they're not slanderous.
I know that if I was slandered I'd be quietly fuming and instructing lawyers, not posturing on twitter and taking pictures of myself 'back in the office' but then again I'm not a tech sector CEO and they seem to be made of different stuff.
What do they say? If the facts support you, pound the facts. If the law supports you, pound the law. If neither support you, pound the table.
"Building different realities" only works to the extent that you surround yourself with sycophants. As soon as the people near you no longer care to play make believe, your alternate reality will *poof* away. Revealed to be nothing but smoke and mirrors.
> Not issuing the statement would be putting a lot of potential liability on them.
They need to proove that Sam wasn't candid once Sam dragged them into an actual court room. Shutting up might not be the winning move in the court of public opinion, but legally, it's probably not a bad move.
Well unfortunately a lot of the tech sector seem to have adopted this weirdly public style of communicating both with each other and in general, in which things that should be discussed quietly in meetings are instead 'discussed' via twitter, including firings and publicly disparaging former employees, which is just tacky tacky behaviour and when you're then forced to publicly retract your statements/action you just end up looking like a complete moron, which is never a good look for a CEO.
It's all very odd really, grown men acting like children online.
The typing style, too, doesn't give a good impression. The weirdly basic language, limited vocabulary, lack of capitalisation - it really gives off a very odd vibe, very unprofessional.
I have literally never seen an announcement by a board when they fired an exec that was as pointed as the OpenAI press release without there being some major malfeasance.
When there is just a standard "disagreement of direction", there are usually some platitudes saying the exec is pursuing other opportunities, or even more forthright saying there was just a difference of opinion.
I think any reasonable observer who is familiar with corporate press releases would naturally assume Altman had done something egregious given the language in the announcement.
That's true, but we cannot fully exclude this just being a stupid, rushed move by the board, either, at least not yet. I would not bet on it, but it's a possibility.
I think you've misunderstood my post. Yes, I do think it was a stupid, rushed move by the board.
My point being is that I think it's totally reasonable for Altman to demand a statement saying there was no malfeasance given that the board's press release was unnecessarily pointed and harsh given that (as all the reporting I've read up to this point, including internal OpenAI memos) it was basically a difference of opinion in the direction of OpenAI that led to his ouster.
> (as all the reporting I've read up to this point, including internal OpenAI memos) it was basically a difference of opinion in the direction of OpenAI that led to his ouster.
The reporting as I've seen has been essentially all one-sided, by people who support Altman, who obviously has a vested interest in portraying him as doing nothing wrong. Even the OpenAI memo doesn't exonerate him of wrongdoing--it just says that any wrongdoing he did doesn't implicate any work going on at OpenAI.
If, for example, Altman was failing to disclose conflicts of interests to the board, that would comport with all statements given so far and still be quite serious wrongdoing for a CEO. Given that the statement was pointed and harsh enough for there to need to be wrongdoing to fully break the board's trust in the CEO, and there's a variety of wrongdoings that Altman could have done that wouldn't contradict available evidence, I'm inclined to believe that he in fact did do one of those wrongdoings.
Hard disagree. Even if you ignore everything that has been reported about the rationale behind the decision, the board objectively executed this decision extremely poorly, managing to piss off their biggest backers, prominent former board members, and lots of their employees. If the board has some evidence of a big smoking gun of egregious behavior, I have to believe they would have shown their cards by now, instead of reports about them bringing Sam back and all of them resigning as terms of his reinstatement.
> I have to believe they would have shown their cards by now
Why? Also their 'cards' are probably quite strictly confidential information. They gain nothing by going to the 'court of public opinion'. They might share their reasoning with some of their largest investors in private, but their lawyers will absolutely be telling them not to reveal anything to the public.
> reports about them bringing Sam back
There are no reports about _them_ bringing him back. There are reports floating about which seem to suggest they will all snap quit and then he will return, but these seem to have no basis and none of them are emanating from the direction of the company. They seem highly speculative, and seem, to me, very much like wishful thinking from the SV 'in-crowd' that all back each other up no matter what.
I do not believe that Sam will return to OpenAI in any role whatsoever, and I believe that the board will hold their position.
I see the board here acting like adults - releasing a professionally written press release then falling silent - while the other side weirdly play to the crowd, not realising that the crowd don't matter, this isn't a reality TV show, the audience doesn't get a vote.
2. Besides the widely reported news about there being discussions, "5PM deadlines", etc. about reinstating Altman, someone obviously let him in the building in an official capacity, and given all the news I highly doubt it was so he could pack up his office on a Sunday: https://twitter.com/sama/status/1726345564059832609 and https://twitter.com/jasonkwon/status/1726348678116548979 (second tweet posted by OpenAI's Chief Strategy Officer).
> None of your speculation comports with the facts
> The OpenAI COO already sent a memo saying Altman was not fired for malfeasance
By what means have you ascertained that the contents of the memo are factual?
> someone obviously let him in the building in an official capacity
Probably one of the same personality-cult types currently carrying water for him. Doesn't mean shit that he got someone to let him in the office. (Probably means they're getting - deservedly - fired.)
Yeah, my immediate assumption upon reading the statement was that there was some scandal far worse than what's revealed here. If there's nothing further, they really screwed up with the wording. They should've fired him as CEO but allowed him to stay on as an advisor and then mutually agreed on some "Sam is resigning to devote more time to his other pursuits" wording, like what happens when most other executives get fired. In this case it'd even be true.
>I would've thought that if he'd done something actually bad they would've come out and said it already
Seems to me that if OpenAI is serious about building AGI, it's reasonable to replace their CEO even if the CEO has done a number of mildly-bad things. We should have high standards for a position that comes with so much responsibility.
He took a rare example of a mission focused and open source tech organization and turned it into just another proprietary silicon valley cash-grab grift.
That alone should be reason enough to fire him and not look back.
Supposing for the sake of argument that the majority of the board privately believe that Sam is a snake, but there's no smoking gun. At this point you're saying they have 2 choices:
1. Let him leave with 60% of the staff. 40% stay at OpenAI
2. Let him win this fight, he's back at OpenAI with de facto 100% control, no check on his power over the AGI that OpenAI is supposedly creating
I'd rather have a snake lead 60% of the staff than 100% of the staff.
On the other hand, if he leaves with 60%, maybe he decides he needs revenge at all costs, and his new AI org doesn't even pretend to be in it for humanity.
Tough choice. I guess we just have to hope that if the board's statement was right, the silent majority of the staff are in agreement.
I’ve seen nothing that indicates this is actually occurring other than rumor laced leaks to tech media. I will wager $20 Altman is the confidential source and the OpenAI board is spending Sunday with their family getting ready for work next week.
If the OpenAI board didn’t spend the entire weekend in a board room figuring out what to do to salvage the organization, then they are a dimwitted as they look right now.
It’s one thing to not have to ask your major benefactors for input when you fire your CEO. It’s another to not do it and expect any further investment, ever. They should be in full on crisis mode and be trying to figure out how to get stakeholders back on board after so publicly slapping them in the face.
They have one of the most valuable products in tech. They have a binding agreement with Microsoft and MS needs OpenAI more than the other way around. They have a dedicated set of researchers who have not pledged to leave and ultimatums have come and gone without mass resignations.
They fired their CEO. It happens. They aren't trying to "salvage" anything. Yes, the for-profit subsidiary investors are probably upset, but the major shareholder is the non-profit who... and this is important... is not beholden to the other for-profit investors!
I don’t fear for OpenAIs future without Altman. I don’t fear they could survive just fine without Microsoft, but it’s unclear the other way around is true. What do people use azure for other than exchange? How much have they gone long bing chat? Do you really think there aren’t deep pockets to back OpenAI? I’d also note OpenAI is profitable as is.
>If the OpenAI board didn’t spend the entire weekend in a board room figuring out what to do to salvage the organization, then they are a dimwitted as they look right now.
Imagine what sort of rubes would fire an employee and then go home afterward. What sort of charity organization doesn’t spend the weekend feverishly F5ing Bloomberg tweets about how disappointed the fired guy is about the situation?
These are exactly the sort of dimwits that would dump somebody and then like, not dedicate their time responding to every text from their exes buddies. Rookie shit.
> I will wager $20 Altman is the confidential source and the OpenAI board is spending Sunday with their family getting ready for work next week.
I'm not going to say it's specifically Altman himself that's the source, but it's been pretty clear to me that all of the sources have been from pro-Altman sources, and in many cases seem pretty clearly directed to increase pressure on the board to accept a counteroffer. Adjusting that for bias... I haven't seen any real indication that the board is actually seriously considering these offers.
I'll also note that it raises my eyebrows that there are apparently two consecutive "accept by X or we quit" deadlines that have now apparently blown by... which is a sign of the side currently in the desperately-negotiating position and which one isn't.
Ding ding ding. Threaten mass resignations if the deadline isn't met... deadline goes by... no mass resignations and the world hasn't fallen. Bluff called.
Sam's push to commercialize and lockdown infrastructure is not in any way incompatible with the mission of OpenAI.
Just as no one is ready for marriage or kids until it happens, we aren't ready for AGI until it happens. It will never be safe enough.
Sam's strategy is to get it out as widely as possible, with incremental progress acclimating us to it. In exchange, OpenAI gets real data on safety over time.
The board is naive and idealistic. Ilya is a genius and his heart's in the right place, but unfortunately his brain isn't.
I love all of these breaking news morsels from the perspective that OpenAI didn’t just fire Sam Altman but rather are struggling through several rounds of trying to hire him, it has made Bloomberg, WSJ, FT etc. sound very smart
OpenAI is one of those unique tech companies where the mantra "build it and they will come" truly applies. While I have nothing against Sam Altman, it's weird to see the significant media attention focused on the CEO. Considering the nature of OpenAI's product, it seems their success might be less about the management and more attributable to the research/technical teams.
This is almost never true unless it's a life saving drug. Xerox literally had the keys to the kingdom for years before multiple people came along and figured out how to get the "tech" to mass market.
This is one of those rare occasions in my opinion. OpenAI trained truly ground breaking models that were miles ahead of anything the world had seen before. Everything else was really just a side show. Their marketing efforts were, at best, average. They chose to call their premier product "ChatGPT", a term that might resonate with AI scientists but appears as a random, forgettable string of letters to the average person. For a considerable period, there was no mobile app available. Their web app had its share of bugs, while third-party apps delivered a better UX. Maybe Sam Altman deserves credit for attracting talent and capital, I don't know. But it seems to me that OpenAI's success by far and large hinges on their game-changing models.
It's interesting to see people comparing Altman to Steve Jobs. By Sam's age, Steve had created the Apple II, created the Macintosh, founded NeXT, taken over Pixar, the list goes on. He was universally recognized as having a great eye for product design as well as being a master marketer who was really good at selling things.
Sam on the other hand founded one startup where he was given $40M and failed. Then he got a job from Paul Graham and did a good job running an established VC firm for 5 years.
He hasn't to my knowledge built any products or run any product companies or sold a bunch of stuff, he's basically a professional investor.
The point here is not to throw shade on Sam Altman who is obviously a bright and successful guy and I'm sure he helped OpenAI get to where they are now. But it's beyond me why there's an army of Altman fanbois on Twitter frothing at the mouth over how this guy is indispensable and the next Steve Jobs and a product/research company like OpenAI can't survive without him. It's all politics/entertainment/rich guys throwing money away basically. It feels like that's all the current incarnation of the Valley is sometimes.
I'm pretty sure the work he put in to brand himself so that there are an army of fanbois bought into his own effort to hype himself up is probably a part of the reason for his firing.
100 comments
[ 1.4 ms ] story [ 185 ms ] threadYou're too used to this tech-sector 'put everything on twitter rn' bullshit immature behaviour.
Do you see the boards of GM, IBM, etc slapfighting on twitter? No, you don't, because it's abnormal.
Totally normal for a board to release a press release and then STFU. Weird twitter slapfights between CEOs and boards are not what we want to encourage. This isn't the pro-wrestling circuit.
Not issuing the statement would be putting a lot of potential liability on them.
Of course, tho whole situation is a little shocking so maybe all bets are off.
"Mr. Altman’s departure follows a deliberative review process by the board, which concluded that he was not consistently candid in his communications with the board, hindering its ability to exercise its responsibilities."
And presumably this is why they used the "not consistently candid" weasel-wording. For that to be true, they only need to point to one material action that he didn't inform the board about (perhaps his chip startup, or Developer Day, or API access), and then point to the board meeting minutes, and then say "See, it's true. You did not inform us, the board, about these other actions you have taken, they matter to OpenAI's adherence to its mission, and therefore your inaction hindered our ability to exercise our responsibilities."
Everything else in the statement appears to be completely factual. It's a bad statement because it unnecessarily burns bridges and implies potential wrongdoing (because every other executive departure statement is so tame), but it didn't, strictly speaking, say anything untrue.
https://openai.com/blog/openai-announces-leadership-transiti...
I agree that in a casual context, this would be mild, but we're talking about professional press releases and judges acting oblivious to this context is not a good thing in my book.
they get it because the other party is afraid of the consequences if they don't get it.
Libel isn't really an issue here, because the statement from the board hardly made any factual assertions in the first place:
> Mr. Altman’s departure follows a deliberative review process by the board, which concluded that he was not consistently candid in his communications with the board, hindering its ability to exercise its responsibilities. The board no longer has confidence in his ability to continue leading OpenAI.
There are a ton of possible interpretations of "he was not consistently candid".
However, there are lots of things that people can sue for besides outright libel/slander. I made the argument elsewhere that the legal nitpicking here isn't really that significant, because it's pretty easy to argue that the tenor of the board announcement would at least make it sound to a reasonable observer that Altman did something egregious.
On HN and twitter people read that line and interpreted it to mean "he's a liar" or even "he raped his sister", but the board didn't actually say those things. He could try suing the board but I think he'd lose.
I haven't read them, but if they're accurate then they're not slanderous.
I know that if I was slandered I'd be quietly fuming and instructing lawyers, not posturing on twitter and taking pictures of myself 'back in the office' but then again I'm not a tech sector CEO and they seem to be made of different stuff.
What do they say? If the facts support you, pound the facts. If the law supports you, pound the law. If neither support you, pound the table.
Because with the capital that good perception affords, you can build a different reality.
Ergo, Altman seems to be spending more effort on managing perception than reality.
Because with capital, you can hire (mercenary) engineers.
As in using money, to pay engineers, to solve the problems you have.
There are fine lines between "fraud", "fake it till you make it", and "projecting confidence".
They need to proove that Sam wasn't candid once Sam dragged them into an actual court room. Shutting up might not be the winning move in the court of public opinion, but legally, it's probably not a bad move.
"If you retract your statement, I won't sue."
"Ok, we retract it."
Just like that.
It's all very odd really, grown men acting like children online.
The typing style, too, doesn't give a good impression. The weirdly basic language, limited vocabulary, lack of capitalisation - it really gives off a very odd vibe, very unprofessional.
When there is just a standard "disagreement of direction", there are usually some platitudes saying the exec is pursuing other opportunities, or even more forthright saying there was just a difference of opinion.
I think any reasonable observer who is familiar with corporate press releases would naturally assume Altman had done something egregious given the language in the announcement.
My point being is that I think it's totally reasonable for Altman to demand a statement saying there was no malfeasance given that the board's press release was unnecessarily pointed and harsh given that (as all the reporting I've read up to this point, including internal OpenAI memos) it was basically a difference of opinion in the direction of OpenAI that led to his ouster.
The reporting as I've seen has been essentially all one-sided, by people who support Altman, who obviously has a vested interest in portraying him as doing nothing wrong. Even the OpenAI memo doesn't exonerate him of wrongdoing--it just says that any wrongdoing he did doesn't implicate any work going on at OpenAI.
If, for example, Altman was failing to disclose conflicts of interests to the board, that would comport with all statements given so far and still be quite serious wrongdoing for a CEO. Given that the statement was pointed and harsh enough for there to need to be wrongdoing to fully break the board's trust in the CEO, and there's a variety of wrongdoings that Altman could have done that wouldn't contradict available evidence, I'm inclined to believe that he in fact did do one of those wrongdoings.
Why? Also their 'cards' are probably quite strictly confidential information. They gain nothing by going to the 'court of public opinion'. They might share their reasoning with some of their largest investors in private, but their lawyers will absolutely be telling them not to reveal anything to the public.
> reports about them bringing Sam back
There are no reports about _them_ bringing him back. There are reports floating about which seem to suggest they will all snap quit and then he will return, but these seem to have no basis and none of them are emanating from the direction of the company. They seem highly speculative, and seem, to me, very much like wishful thinking from the SV 'in-crowd' that all back each other up no matter what.
I do not believe that Sam will return to OpenAI in any role whatsoever, and I believe that the board will hold their position.
I see the board here acting like adults - releasing a professionally written press release then falling silent - while the other side weirdly play to the crowd, not realising that the crowd don't matter, this isn't a reality TV show, the audience doesn't get a vote.
1. The OpenAI COO already sent a memo saying Altman was not fired for malfeasance but for a "breakdown in communication" between the CEO and the board: https://www.axios.com/2023/11/18/openai-memo-altman-firing-m...
2. Besides the widely reported news about there being discussions, "5PM deadlines", etc. about reinstating Altman, someone obviously let him in the building in an official capacity, and given all the news I highly doubt it was so he could pack up his office on a Sunday: https://twitter.com/sama/status/1726345564059832609 and https://twitter.com/jasonkwon/status/1726348678116548979 (second tweet posted by OpenAI's Chief Strategy Officer).
> The OpenAI COO already sent a memo saying Altman was not fired for malfeasance
By what means have you ascertained that the contents of the memo are factual?
> someone obviously let him in the building in an official capacity
Probably one of the same personality-cult types currently carrying water for him. Doesn't mean shit that he got someone to let him in the office. (Probably means they're getting - deservedly - fired.)
Do it with an agreement that he waives his right to sue.
Seems to me that if OpenAI is serious about building AGI, it's reasonable to replace their CEO even if the CEO has done a number of mildly-bad things. We should have high standards for a position that comes with so much responsibility.
That alone should be reason enough to fire him and not look back.
No matter though. Open source alternatives are catching up in a fraction of the development time. OpenAI has been left behind in the race for Open AI.
Probably would be the best thing that could happen in turns of actually making AI "open", but also would unleash utter mayhem in the marketplace.
https://www.semianalysis.com/p/google-we-have-no-moat-and-ne...
1. Let him leave with 60% of the staff. 40% stay at OpenAI
2. Let him win this fight, he's back at OpenAI with de facto 100% control, no check on his power over the AGI that OpenAI is supposedly creating
I'd rather have a snake lead 60% of the staff than 100% of the staff.
On the other hand, if he leaves with 60%, maybe he decides he needs revenge at all costs, and his new AI org doesn't even pretend to be in it for humanity.
Tough choice. I guess we just have to hope that if the board's statement was right, the silent majority of the staff are in agreement.
Edit: I hope Trent Reznor does the soundtrack!
It’s one thing to not have to ask your major benefactors for input when you fire your CEO. It’s another to not do it and expect any further investment, ever. They should be in full on crisis mode and be trying to figure out how to get stakeholders back on board after so publicly slapping them in the face.
They have one of the most valuable products in tech. They have a binding agreement with Microsoft and MS needs OpenAI more than the other way around. They have a dedicated set of researchers who have not pledged to leave and ultimatums have come and gone without mass resignations.
They fired their CEO. It happens. They aren't trying to "salvage" anything. Yes, the for-profit subsidiary investors are probably upset, but the major shareholder is the non-profit who... and this is important... is not beholden to the other for-profit investors!
You think any agreement with Microsoft doesn’t have outs?
Imagine what sort of rubes would fire an employee and then go home afterward. What sort of charity organization doesn’t spend the weekend feverishly F5ing Bloomberg tweets about how disappointed the fired guy is about the situation?
These are exactly the sort of dimwits that would dump somebody and then like, not dedicate their time responding to every text from their exes buddies. Rookie shit.
Altman was posting that he's at openai. So, while media reports might be posturing/inaccurate, it's very likely there's some negotiating going on. https://twitter.com/sama/status/1726345564059832609
I'm not going to say it's specifically Altman himself that's the source, but it's been pretty clear to me that all of the sources have been from pro-Altman sources, and in many cases seem pretty clearly directed to increase pressure on the board to accept a counteroffer. Adjusting that for bias... I haven't seen any real indication that the board is actually seriously considering these offers.
I'll also note that it raises my eyebrows that there are apparently two consecutive "accept by X or we quit" deadlines that have now apparently blown by... which is a sign of the side currently in the desperately-negotiating position and which one isn't.
Just as no one is ready for marriage or kids until it happens, we aren't ready for AGI until it happens. It will never be safe enough.
Sam's strategy is to get it out as widely as possible, with incremental progress acclimating us to it. In exchange, OpenAI gets real data on safety over time.
The board is naive and idealistic. Ilya is a genius and his heart's in the right place, but unfortunately his brain isn't.
Sam on the other hand founded one startup where he was given $40M and failed. Then he got a job from Paul Graham and did a good job running an established VC firm for 5 years.
He hasn't to my knowledge built any products or run any product companies or sold a bunch of stuff, he's basically a professional investor.
The point here is not to throw shade on Sam Altman who is obviously a bright and successful guy and I'm sure he helped OpenAI get to where they are now. But it's beyond me why there's an army of Altman fanbois on Twitter frothing at the mouth over how this guy is indispensable and the next Steve Jobs and a product/research company like OpenAI can't survive without him. It's all politics/entertainment/rich guys throwing money away basically. It feels like that's all the current incarnation of the Valley is sometimes.