OpenAI's board approached Anthropic CEO about top job and merger | Hacker News Reader

OpenAI's board approached Anthropic CEO about top job and merger

reuters.com

88 pointsdongping75 comments

Looking more and more like the dev day nuking of D’Angelo’s Poe led to this outcome. I’m surprised that connection has escaped close scrutiny - it was a conflict of interest AND now Poe is viable again as intended. For a minor player in AI he’s managed to maximize his interests and while serving on a non-profit board.

This theory is endorsed by OpenAI's Head of Applied Research, Dr. Boris Power, who retweeted a Robert Scoble tweet saying this. See his twitter timeline here https://twitter.com/borismpower?lang=en

For those not logged in, Power said the following was "an interesting thread, maybe too speculative":

https://nitter.net/scottastevenson/status/172673102286200873...

Correct, and Dr. Power also retweeted a tweet from @Scobleizer that says:

    Who held the gun that killed OpenAI?

    Hey 
    @adamdangelo
     all evidence points at you.

    Wanna come on my show?

Being spun as deeply ideological seems about right - the 3 non-profit oriented board members seem like they were already open about their views in that direction, so just took one vote to flip - D’Angelo.

I see stuff like this spreading in social media and it’s always people speculating without any evidence. On Saturday it was a coup by Ilya, on Sunday it was a coup by effective altruists/decelerationists on the board, today is D’Angelo’s Poe getting eaten up tomorrow it will be something else.

Your theory collapses in the slightest amount of scrutiny, if D’Angelo tries to destroy the company due to his own company losing value why would the other 3 board members be on board with that? You’re not doing anyone any good speculating stuff like this just adding more noise.

Not saying I agree with the theory, but if D’Angelo was any but manipulative he would obviously lie and convince them using other reasons. For instance he could claim that he’s worried about alignment, safety, etc.

any *bit manipulative

Your argumentation against parent is very fluffy.

> Your theory collapses in the slightest amount of scrutiny, if D’Angelo tries to destroy the company due to his own company losing value why would the other 3 board members be on board with that?

The other 3 have their own motives which have been repeated over and over here. It came to light that there were strong tensions between Helen and Sam as he wanted her to be removed from the board. Ilya was hurt by Sam over the competing teams fiasco. Adam wanted to protect his competitor which is a clear interest conflict and could bring him in severe legal trouble.

And you figured out 4 different motives for 4 different people based on what exactly?

The motives are well documented. Whether these motives were enough to pull the trigger, or other motives will appear remains to be seen.

You don’t think D’Angelo has a conflict of interest?

One is not enough you need 4

Re: "destroy the company"

It's plausible that none of the key participants had really thought this through or imagined that the co would lose value as a result.

I think you might have mixed up effective accelerationist with effective altruists there, but I couldn't say for sure

>On Saturday it was a coup by Ilya, on Sunday it was a coup by effective altruists/decelerationists on the board, today is D’Angelo’s Poe getting eaten up tomorrow it will be something else.

It's already moved on to Helen Toner.

https://twitter.com/karaswisher/status/1726755613667934217?s...

>>why would the other 3 board members be on board with that?

I guess all had their reasons. This is one big reason why you don't want these non-profit kind of things. No stake, No play. Human beings don't feel about the things they don't own the same way, they feel about things they own.

I think the motivation to make profits, get rewarded is a tremendous driver of things. In situations like these you are always tempted to think that the true end of altruism is nobody(who works on things to make them happen) benefits. Notice how its the same when the set up succeeds, same when it fails.

If D'Angelo acted to protect his company it shouldn't be surprising. The whole point of having people with conflict of interests to not serve on your board is exactly this. They have more reasons to see you fail, than see you win.

Other people likely had their own reasons. Ilya probably was acting too altruistic, and had thoughts about Sama for a long time.

Expecting the whole company to sign up with your moves, especially when their financial interests are the exact opposite, is taking it a little too far. Which again brings us back the same thing. Lots of people seem to be arguing that people working on this at all levels must be doing this for grander reasons, and not making it about just money alone. Which is so unfair and such wrong evaluation of people's motives to work on anything at all, let alone this.

Finally, it feels like nobody wanted to see the company win. And they didn't have incentives to do that either. Every one seems to be happy it ended.

D’Angelo is certainly naive in his self-interest, and definitely not capable as to orchestrate this thing. Think of him like he's useful bastard, and now it's his turn to become the scape-goat bastard for everybody to hate. This is counter-spin for containment purposes. The real questions we should be asking is how come the NATSEC patriot-ladies on the board managed to exploit Sutskever in such blatant way, what that entailed, and what kind of psychosis it took to get there.

It has not escaped close scrutiny. It's just that people don't go around accusing a specific person without evidence.

The timing of Dev Day and its impacts on Poe aren’t suspect to you?

And you don’t think D’Angelo has a conflict of interest that impacts his board role at OpenAI?

This will be a fun book to read after it's all done.

As its happening I had no idea what is going on though. Surely the board can not be this incompetent? Who kicks out a current CEO without having found a replacement or at the very least a shortlist. No evidence has come out that sama did anything egregious against OpenAI. Truly one of the most bizarre sequence of events I have ever witnessed.

I believe they had a shortlist, but no one stepped up to take the job.

Even then, that “shortlist” was clearly built in a total scramble after their first interim CEO sided with Sam.

There has to be a good reason on why this was rushed, it has to be a "now or never" kind of situation for this outcome. I presume the candidates on the shortlist were skeptical to accept the job until the smoke clears on the situation, they do not want to be on the bad side of this affair until it is known that accepting the job is not a career and network suicide.

Which is at least good to hear but typically you would think there would have been feelers out there to gauge temperature. Perhaps this is indicative of a board that does not have as much business professional experience.

Still hilarious that Emmett Shear was not only offered the role but accepted. Amusing to see the individual who said we need to severely slow down AI work to now be heading up the leading company in the space.

Edit: I guess the true measure is that nobody read the room from employees and went ahead with this whole fiasco.

Almost every story we get seems to point in the direction of the board being erratic and incompetent. I feel like either that has to be the obvious truth, or I am being gaslighted by a concerted effort from the other side.

I also noticed an unusually high amount of throwaway accounts on OpenAI threads lately. I am not sure whether it's a concerted effort but Altman does seem to have a lot of fans.

Rather than putting up a conspiracy theory the much more plausible explanation is that this board evidently pissed off a lot of customers, powerful people in tech and almost all employees of openai.

If we are to believe media reports. But why do customers and employees care that much about having Sam Altman as CEO? Sam was just at the right place at the right time. He is no doubt competent but not some kind of irreplaceable AI or management prodigy. OpenAI's true strength lies in its AI research team and their groundbreaking models. The C-suite is just a sideshow.

This is what I’m confused about. There are plenty of product people who would have been chomping at the bit to release ChatGPT if they knew their organization had the tech. I’m not sure what makes him unique here.

He is a known quantity who has been successful leading the company. But of course he could be replaced and there are others that could be equally, maybe more successful.

The problem is not that Sam has some totally unique DNA. Customers are reacting to 1) the board’s apparent rejection of the existing products and business model, and 2) the capricious and incompetent way the board handled this.

Companies are making huge bets, and a seemingly stable and industry-leading supplier just turned out to be unreliable. Of course people are angry and hesitant to keep doing business with OpenAI.

I'm sure this question can be answered better by the employees themselves and they voted very clearly that they don't want a leadership unter the current board that ousted Sam and Greg. As for why customers care about who the CEO is, i'm not even sure if you're seriously asking these questions.

I see your point, but I am somewhat skeptical about leaks. I'm open to being convinced about how Sam's leadership played a pivotal role in OpenAI's success, or why finding a suitable replacement might be difficult. Are there specific instances where his decisions clearly influenced the course of events?

Observing from the outside, it feels like the bulk of the credit goes to their AI team and that Sam was just there to make sure the machine was well oiled. Maybe he was really good at that, I don't know. Their productization of models wasn't that great, starting with the name, "ChatGPT". IMO, the real driving force has been the unparalleled capabilities of their models rather than branding or marketing.

> Sam was just there to make sure the machine was well oiled

That's a big part of the success of any company. Sam and Greg were responsible for recruiting the team that made it possible and removing the many obstacles those people faced over the years, even though their competition had much more money and prestige at the starting point.

As a customer/potential developer I don't care who the CEO is. However, I do wish the company was stable and the current situation seems anything but. The employees' letter makes me anxious about the company's future.

makes sense.

It seems this was a coup of Microsoft together with Altman to go full force ahead with the commercial upscaling.

In the past there was already a small uprising, and all who didn't like the Altman cult ran off to anthropic. So all that's left in openai are huge fans of Altman. So it only makes sense that you piss off almost all remaining employees, its survivor bias at work.

Some people on Twitter, notably @HackingButLegal, have been collecting examples of possible paid astroturfing accounts which she believes belong to a reputation management firm hired by Sam to suppress his sister Annie's accusations that Sam sexually abused her as a child.

https://twitter.com/HackingButLegal/status/17268410084771104...

https://twitter.com/HackingButLegal/status/17268395129057322...

Have you actually read the sister's accusations? They read like the rantings of a paranoid schizophrenic. One or two things, I could believe, but the sheer volume of ridiculously extreme accusations is impossible to believe.

Her core accusation is that Sam would always read her bedtime stories (he was almost a decade older) while laying in bed with her, and that sometimes while they were in bed he would cuddle her inappropriately. And that at some point she started asking him not to get in the bed and he ignored her “no” and kept up the nightly ritual.

This seems well within the realm of possibility. She probably does have mental health issues (maybe paranoid schizophrenia), but that could be an effect of the childhood sexual trauma. I had a 40-something-year-old aunt who committed suicide due to being raped as a child. :( Early trauma like that can fuck you up in ways that make you permanently unstable.

Getting abused will mess you up in all sorts of ways.

And in any case, someone as rich as Sam Altman letting a close relative live in poverty like that is by itself a pretty damning inditement of his character.

Would you feel better had he locked her up in premier upclass mental institution?

You don’t see any other possibilities? Not, say, paying her rent (or, realistically, buying a house with his spare change and letting her live there) or chipping in for therapy? His net worth is estimated to be in the half billion range so he’d never even notice the cost.

I'm not prepared to assess people's personal business, and I believe it's inappropriate for us to discuss their relationship as none of us are privy to their personal history, who paid for what, who approached who or approached not, and so on.

This thread only exists because you did feel it was appropriate to comment. My point was simply that you assumed one of the worst options rather than something more charitable.

Google says the guy's worth $500M. A mere 1% of that in a trust could ensure that none of his siblings would have to live in poverty ever again. In absolute numbers that's a lot, but in relative terms I've given away much larger percentages of my net worth to relatives, including ones more distant than siblings, I'd do it again, and I'd think less of anybody who wouldn't do the same.

There are so many pro-Sam posts from well established and even noteworthy accounts. What would be the point of astroturfing a majority opinion?

As long as we’re throwing around evidence-free accusations of bad faith, the astroturfing claims sure look like the age-old tactic of claiming anyone who disagrees must be a shill.

I feel called out - I have created a HN account recently in order to comment on this matter.

I consider the recent events as crucial to the future of whole industry. I follow them extensively and I am greatly interested in participating in a conversation with informed people. What better venues are there? Twitter? Reddit? Those two platforms are sadly full of uninformed trolls.

What's erratic here? If they are likely to lose most of their employees and resources then merging with the most aligned other company is not only reasonable but exactly what it says they'll do in their charter.

I don't know, I have no independent means of verifying whether the story is true or not in the first place.

I think it comes across that way because they actually take the non-profit original mission and ethics more seriously than the for-profit business, which is an incomprehensible shock to most capitalists. However, it makes sense if Anthropic left in the first place because they didn't agree with sama's direction of the company, and this is just round 2 of the OAI civil war in regards to whether OAI's original non-profit mission ACTUALLY matters or not.

The board has bylaws governing how it works. One of them says that it has to have 48 hours notice before a meeting (to, e.g. agenda, minutes, resolutions, actions ). I think a board member cannot speak officially before that. We might be an announcement today or tomorrow. Perhaps they have had to have a meeting to get legal advice before an announcement?

I think the board is just being as slow as any board is. But they didn't plan for the somewhat foreseeable response and we are impatient.

This is as charitable as I could think of.

Not to get too conspiracy, but remember where you get your news from: HN, maybe reddit?

Guess which platforms have deep ties to Altman?

This thread just shows why people in tech have 0 social intelligence and instead follows the trending mob fashion of the time.

A writer for wired says they have reason to believe this story is a fabrication (although doesn't explain why):

https://twitter.com/willknight/status/1726793735143621058

The only difference: the informer is a respected publication while wired is wired and you're not gonna make it better with a claim and then not follow-up with any kind of evidence. But at this point it could be all fabrication. Journalism is mostly a joke at this point.

Is The Information actually a higher quality source? I can't say I know much about them. Why do you think that?

A lot of gaslighting, as OpenAI findings are a subject of national security.

Judging from the current situation, openai's board of directors are all shameless people. If they still have dignity, they should resign immediately. They were unqualified and incapable of managing one of the most important technology companies in the world, and they destroyed a shining company, a great product, and the dreams of countless people. I will never use Poe and Quora again, and I will boycott any product related to Adam D'Angelo. Adam D'Angelo is an asshole.He doesn't have any shares in OpenAI, so he doesn't care about destroying OpenAI at all.Absolutely ridiculous, the only reason I can think of is that he's extremely jealous of Sam.

This whole thing becomes more and more like a Schooby and the Gang case. Let's see who's under the mask at the end. :D *Gonna need some popcorn.

Reposting what dnissley said that got downvoted:

Will Knight, a writer for Wired, is stating “I have reason to believe that the OpenAI Anthropic merger story is a complete fabrication”

https://twitter.com/willknight/status/1726793735143621058

> Reposting what dnissley said that got downvoted:

Please don't do this.

https://news.ycombinator.com/item?id=38363524 isn't even downvoted now.

The board is desperate. A $90B company turned overnight into a lemon nobody wants, despite having amazing tech and APIs. They need to find a way to keep at least 20% of the employees from leaving, as that will allow them to continue running the services that have paying customers.

This looks more and more like the UBA merger in The Morning Show.

Some journalists on X are claiming that this story is fabricated

Maybe some links please? I also don't want to share a fabricated story.

This could be counted as a felony against individual board members as there were clear conflict of interests among them with such a merger. I'm sure there will be lawsuits from the investors.

dupe https://news.ycombinator.com/item?id=38357629

Solution: an AI replaces the board.

Can't do much worse.

Netflix should hire whoever is writing the script for this drama. It's gripping!

These people are going to get sued into oblivion.

I’m voting for Dan McCrum, Elliott Brown, Maureen Farrell, Duncan Mavin and Bethany McLean to team up and write the book to end all books about corporate stupidity.

I tried not to have an opinion until now, but so far Altman seems like a greedy asshole, and D'Angelo seems like a greedier asshole.

That clash ended up terribly, since neither of them care about anything else than their best interests.

How is Altman a greedy asshole?

He's been at OpenAI since founding date, he's taken zero economic upside in this venture, and he's now being kicked out - what exactly makes him an asshole?

Zero economic upside for being the OpenAI CEO? That is difficult to believe.

I got this picture from other dirt revealed during this whole debacle: the fact that he wanted to raise billions to run a chip company for AI (obvious conflict of interests), how he lies and manipulates people, even sexual abuse allegations by his sister.

Brockman sounds great, but not sure why the tech world wants to praise someone like Altman. Maybe more people that I would expect what to be like him, even if it means doing some lying and manipulation from time to time.

Reply on news.ycombinator.com