OpenAI Committed to Buying $51M of AI Chips from a Startup Backed by Sam Altman
wired.com
wired.com
In WeWork, the CEO took a trademark that costs a couple of thousand dollars, and charged himself millions for it from company money - that's personal enrichment. Incubating a thing considered highly risky and expensive 5 years is not personal enrichment, its personal risk that helps more the likelihood of it being less impossible instead of a chance at enrichment from two companies you don't even know would be viable (OpenAI) 5 years from then.
"OpenAI does not have to buy the chips unless they meet the criteria,"
The second sentence suggests if the chips meet the criteria then the LOI is binding.
Even if, for example, there were other chips from other companies, in which Altman held no interest, that met the criteria.
The phrase "have to buy the chips" suggests more than simply a requirement to negotiate in good faith.
But if we cannot see the text of this LOI then we are all just speculating.
The odds of success of that 1M investment and the chip company were closer to 0 than 100%
The odds of success of openai in a competitive environment with massive monopolies who were building AI was closer to 0 than 100%.
Trump was president. He was rattling the chains of every trade partner and China was threatening the most important chip-making places.
Sam altman was co-chairman of OpenAI, not CEO, and still led YC.
There could be plenty of other things about him to be concerned about, but this investment, with the information we have about that time, is not. Especially because you don’t get AI without chips.
The board lost trust for his honesty from many small things that kept accumulating. Altman is back on condition that there is a investigation.
But considering the board initially fired him for being dishonest/lying in an ongoing fashion, and the board wouldn't do the firing without evidence they could back it up with, it's not so hard to extrapolate that a return would be conditional on a deeper investigation and "full" cooperation.
Where's the evidence then? Why did they reinstate him if they had such amazing evidence of wrongdoing?
Since then, journalists have talked to both sides on and off the record and confirmed that the only reason for firing was distrust for Altman's honesty.
Altman's history seems to match. https://www.washingtonpost.com/technology/2023/11/22/sam-alt...
>Four years ago, one of Altman’s mentors, Y Combinator founder Paul Graham, flew from the United Kingdom to San Francisco to give his protégé the boot, according to three people familiar with the incident, which has not been previously reported. .... he flew across the Atlantic with concerns that the company’s president put his own interests ahead of the organization — worries that would be echoed by OpenAI’s board.
>A separate concern, unrelated to his initial firing, was that Altman personally invested in start-ups he discovered through the incubator using a fund he created with his brother Jack — a kind of double-dipping for personal enrichment that was practiced by other founders and later limited by the organization.
“It was the school of loose management that is all about prioritizing what’s in it for me,”
[1] https://x.com/geoffreyirving/status/1726754277618491416?s=20
It sounds a lot like background briefing to me...
"So committed I can't be honest"??
If you can't be honest and inform people of possible conflicts of interest, it is bad and you know yourself it's bad.
Yes, “IF” you can’t inform people of possible conflicts of interest, that is bad. But it’s also irrelevant as of yet.
I see nothing wrong here.
Ironically, your follow up comment does point out the typical HN rush to judgment, a much more literal ad hominem.
> So they can shrink from and deny any possible conflict.
you quite literally insinuated negative motive and action to follow directed towards an individual using a pronoun. take some personal responsibility.
Is there another reason you were joking instead of just making your point?
OpenAI isn't a government agency or entity. They're free to no-bid or to use and purchase resources that someone within the company has ties to. The replies here are loaded with accusations of fraud; A typical HN rush to judgement.
Having investors, even if you own a controlling interest, still has implications on company decisions and possible legal action.
1. Many people would still scream bloody murder because they do not like Elon
2. When it comes to Tesla it is a public company so Elon has a duty of care to the public shareholders
3. even for the Private companies, Elon is not the SOLE owner, not even of twitter, so he still has a duty to the other minor shareholders, being a majority shareholder gives you lots of authority, but not unlimited authority. If he was the Sole shareholder it would be different.
Can't tell if this is meaningless, the reporter just failed to convey its actual meaning, or I am too dumb to understand
It's definitely both of the first two.
Generally I wouldn't gatekeep, but here, it seems apt.
The American education system is no measure of a person's worth. I've already built and sold one startup somehow despite your sneering, and I dropped out after a semester and a half. I'm now working on my second. Try to avoid the urge to casually dismiss entire categories of people because you want to elevate yourself above one of them. Books are available without spending $40k per year.
If you don't know what Worldcoin is, look it up. It doesn't really need any commentary.
The fact that this dude was basically handed YCombinator and then handed OpenAI shows how incestuous the inside-most track of Silicon Valley is. I guess it's pretty much identical to Hollywood.
It's like that for every billionaire, humanity will make some X progress in some Y field and some lucky schnook will find themselves in the epicenter of it all.
Just like on the other hand an Earthquake will stuck in coordinates X,Y and some poor unlucky schnook Z will find themselves in the epicenter of it all
There could be conflict of interest or it could also be that OpenAI has no desire to get into the chip business.
The existence of such a conflict is not itself inherently a problem; these situations come up all the time, become disclosed, and extra scrutiny is placed on decisions where the conflict is likely to surface.
I don’t find it at all obvious that a billionaire CEO would improperly influence the company he’s running to buy from a company he’s invested in over a few million in chips. It’s a topic ripe for speculation and insinuation and hopefully the conflict of interest was properly disclosed and someone other than Altman made the final decision to commit to this purchase.
Saying that you are not in it for the money is becoming like a fashion statement. What’s wrong in saying that you’re in it for both money and passion?
One of those things we all know is bullshit but we just pretend not smelling it.
And it doesn't mean the most profitable option will be picked, but it does ensure that easy money is not left on the table, while lots of other money is set on fire in pursuit of the personal interests of director level managers.
Take time to smell the bullshit.
[1] https://www.wakeforestlawreview.com/2012/04/our-continuing-s...
51% ownership doesn't mean you can do anything you want with the company. It's sort of a main point of a LLC.
There are a lot of misconceptions about maximizing shareholder value, even among economists. But talk to a legal scholar or a corporate lawyer: a CEO or board is not legally obliged to maximize shareholder value. They need to maximize the value of the corporation and act in its best interest. Only when there is a change in legal control, such as a merger or imminent hostile takeover, do they have to maximize shareholder value.
Just to be very clear: modern corporate law does not require profits at the expense of everything else, and maximizing profits or shareholder value is not the same thing as serving shareholders’ best interest.
If you disagree with what the officers think is the best way to carry out these duties, the courts say your option is not to sue them, it is to invest in a different company.
Why is it so hard for people to accept that for profit means for profit?
You can't even point to successful companies that treat consequences from violating laws or morality as just a cost of doing business and say that forces everyone else to follow suit. What might be good for one company isn't necessarily good for another.
The fact that companies advertising on twitter don't want their products appearing next to far right diatribe is proof enough that considering morality and ethics is important to maximizing value.
LLCs are amoral by design. It can be argued that this amorality can result in good outcomes when e.g. regulated proprerly. But that doesn't make the LLCs themselves moral actors.
Mistaking companies for moral actors results in folly like letting them regulate themselves.
That is the problem with the modern era, your actions do not matter your words do... Words have become violence, but actual violence is excused.
If you do not say the "correct words" your actions matter not. inversely as long as you say the "correct words" your actions matter not.
It seems anathema, but it isn't, Zen, arguably, was instrumental in the Imperial Army running roughshod over its enemies: https://en.wikipedia.org/wiki/Zen_at_War , so too with Executives -they simply become instruments of fate itself, it's not they themselves doing it.
AI chips are expensive and hard to get; nvidia can't produce enough; nvidia's chips are very expensive; building / seeding alternate supply for a crucial input is just competent business.
Doubt it. This type of thing is very common.
>Altman is back on condition that there is a investigation.
Altman is back because OAI had no choice.
It looks to me like exactly the kind of investment I would expect a wealthy angel to make in startups related to their interests, not undue bias.
I do wonder if this kind of thing was related to his ouster-and-return, but it happened in 2019 so it would need to be a pattern of behavior beyond what we see in the article.
So he has an incentive to do business with a chip maker that he invested in. But that's allowed right? Or is the argument more that it's like insider trading (investing?) because he knew he might announce business with it in the future? Since it's not publicly traded, I don't even know if that's unethical.
It would be very hard to prove that choosing company A over company B was a knowingly worse choice for OpenAI though.
[1] https://www.thestreet.com/investors/sam-altman-net-worth-how...
Also, depending how early he was, he could own significantly more than 5%.
One of those bets is on the fusion power startup Helion Energy, into which he’s poured more than $375 million, he told CNBC in 2021. The other is Retro, to which Altman cut checks totaling $180 million the same year.
“It’s a lot. I basically just took all my liquid net worth and put it into these two companies,” Altman says.
Whoever thinks that was a good idea doesn’t seem to have any concept - nor care for - morality and ethics, in my view.
Maybe the board firing him was exactly over this and look who prevailed.
You can see this with the apple touch ID implementation when you have multiple accounts on the same computer.
[0] I don't, he clearly does
[1] I am not sure, I want to but suspect it will cause inflation unless done very carefully, but UBI needs biometrics to avoid Sybil attacks
The answer is that you don't, and Worldcoin is just another giant ball of shit.
Yeah, that doesn't help decentralization.
Passing audits is necessary, but nowhere near sufficient.
I think it comes down to, is this good for this company. If not and you are directing sales to another company that benefits you, that would be bad, but if buying from the other company you invest in is good for OpenAI, then it's ok.
You are allow to have other investments, it comes down to disclosure, and not making a bad deal for reasons that personally profit you.
I was helping to implement ISO through small orgs (most likely overkill...) and remember one of important bits was that each purchase had to go through a process of selecting supplier. I wonder if this is also the case in OpenAI (I'd guess it is?). But I also can imagine, if there is only one supplier who provides given goods then the process is not very helpful.
You could certainly make an argument for that, but it's not at all the standard practice currently, so I wouldn't say it's fair to single out Sam in this case unless there's other evidence of self-dealing.
At the very least you should let the board handle any potential dealings with the company and have 100% disclosure at all times (as you say)
Again, not a lawyer, but I think there is a legal difference between the appearance of conflict of interest, and an actual conflict.
What this comes down to is: a. Is it legal? b. Is any grey area worth challenging in court, if possible at all? c. OAI is operating under a weird form of public scrutiny that other companies largely aren't, for normal if questionable phenomena like this issue, because the public is suffering under the illusion that they have a company stake (and that somehow decisions like this are indicative of a moral hazard that jeapordizes that stake).
Nothing is going to stop LLM ("AI") re-modeling of the communal world. OAI will be far from the only player in that process. If anything, all of this nonsense attention on Altman's minutia will just serve to distract from anything actually helpful that could be done.
Not sure "making incredible investment choices" is the word I would use.
The problem was their complete silence on what lead to Altman's ousting
But the main point of your comment still stands :-)
"...Some time in the future, and it may be soon, you will be told by someone that the Eskimos have many or dozens or scores or hundreds of words for snow. You, gentle reader, must decide here and now whether you are going to let them get away with it, or whether you are going to be true to your position as an Expert On Language by calling them on it..."
> Among the many depressing things about this credulous transmission and elaboration of a false claim is that even if there were a large number of roots for different snow types in some Arctic language, this would not, objectively, be intellectually interesting; it would be a most mundane and unremarkable fact.
If so, the burden of proof falls on him to prove it's not the case. The whole thing comes off as a fun rant about how it's not such a big deal even if true, while never actually addressing its truthfulness.
How's this for sources? https://en.m.wikipedia.org/wiki/Eskimo_words_for_snow
> Krupnik and others charted the vocabulary of about 10 Inuit and Yupik dialects and concluded that they indeed have many more words for snow than English does. Central Siberian Yupik has 40 terms. Inuit dialect spoken in Canada's Nunavik region has at least 53, including matsaaruti, for wet snow that can be used to ice a sleigh's runners, and pukak, for crystalline powder snow that looks like salt. Within these dialects, the vocabulary associated with sea ice is even richer. In the Inupiaq dialect of Wales, Alaska, Krupnik documented 70 terms for ice including: utuqaq, ice that lasts year after year; siguliaksraq, a patchwork layer of crystals that form as the sea begins to freeze; and auniq, ice that is filled with holes. Similarly, the Sami people, who live in the northern tips of Scandinavia and Russia, use at least 180 words related to snow and ice, according to Ole Henrik Magga, a linguist in Norway.
I find it annoying that anyone would imply it's unintellectual to consider this possibility based on such an intellectually vacuous article as the one you shared.
The alternative was that OpenAI prepaid for these chips, and took on a lot of risk. From Altman's position, he's personally taking on risk instead of having OpenAI (and its investors) take it on.
I mean, one can just say, I don't know and don't care about this and that is their personal problem to deal with while I receive millions of dollars in shares.
The whole point is stay away and don't play the "MeToo" game.
Does anyone get tired by the tech bros label always being thrown out when people disagree. Can we not all grow up and if we are to have disagreements, actually discuss it constructively?
Maybe life has always been like this but I find it harder and harder to have meaningful discourse. If you disagree with someone then all of a sudden you get put into a generalized bucket.
Edit: I don’t like how Elon always handles himself in public and I don’t always agree with everything Altman does but we as humans are very different and that’s ok. The moment Tesla comes up all the Elon hate comes out of their holes.
In this case if the article was about some kind of misconduct or worse towards another person ok great let’s talk about his sister.
Potentially as a result of or exacerbated by abuse, or tolerance thereof? Or made her an easier target?
> it serves no purpose to a a thread about OpenAI buying chips from a company he is invested in.
Certainly it does. Self-interest, conflict of interest etc., raises valid concerns about a person's sense of morality and ethics. Much like alleged assault or abuse of a vulnerable family member.
Edit: Don't even bother replying, this person is a Elon/Altman hater. The majority of comments are just negative spam against any kind of thread related to OpenAI or Tesla.
You're right, in a criminal court. I don't know what the evidence is or isn't. Neither do you. I just posited a theory. The very vast majority of abuse has no "proof" and often boils down to he said, she said.
> Don't even bother replying, this person is a Elon/Altman hater. The majority of comments are just negative spam against any kind of thread related to OpenAI or Tesla.
Let's see, 9,114 comments on HN, 100 or so which relate to Musk/Tesla, and approximately 20 of which reference the OpenAI drama.
Revisit the site guidelines:
* Please don't post shallow dismissals, especially of other people's work. A good critical comment teaches us something.
* Please respond to the strongest plausible interpretation of what someone says, not a weaker one that's easier to criticize. Assume good faith.
- I make, no, I paid enough for health insurance. I have no equity in OpenAI.
- Really? That's interesting.
In 2008: "Bill Gates pledges to leave his £30billion fortune to charity... rather than his children" - https://www.dailymail.co.uk/news/article-1027878/Bill-Gates-...
In 2022: "Bill Gates pledges to donate ‘virtually all’ of $113bn fortune to his foundation" - https://www.theguardian.com/us-news/2022/jul/15/bill-gates-b...
In the meanwhile in 2023 is still the 6th richest. Giving money away is such a hard job :-))
"Global billionaire tax could yield $250 billion annually, study says" - https://www.reuters.com/business/finance/global-billionaire-...
Vertical integration is a competitive advantage worth pursuing for OpenAI. The spin is just that - spin.
That would leave Meta's other shareholders paying the bill while missing out on the gains if the project becomes a success. In other words, the risk is shared but the potential rewards are reaped by just one shareholder.
But, Zuckerberg has done some stuff maybe somewhat similar to the WeWork trademark thing: the "Meta" trademark came from his quasi-charity's investment into some kind of meta-study medical research company. I'm not sure there was any conflict there though as I think Chan-Zuckerberg initiative fully owned it and didn't have other shareholders.
This is instead a question of whether Sama preferentially made a contract with Rain, perhaps despite better alternatives. (Which would benefit him but might not be in the best interests of stake holders).
It’s possible it was the correct decision though, and his stake is just a fortunate side-effect, after all Sam is also invested in other chip companies (Cerebras I believe?).
Good for Sam, I say. Money aside he should be allowed to keep his interests separate. Just because this other company is AI related doesn’t mean it’s working in an area OpenAI was even interested in pursuing.
Just to be clear, I never met the guy and have no affiliation with OpenAI.