Sam Altman is a pathological liar
https://www.theregister.com/2024/05/29/ex_openai_board_membe...
Sam Altman is a pathological liar
https://www.theregister.com/2024/05/29/ex_openai_board_membe...
I only skimmed through the beginning but I am not sure what the news is. Looks like they mostly/only? pulled it from some filings that I am sure was being put together by third party agents.
I am not sure what the news is
The news is that lying on those filings is a federal crime and can lead to jail time (for normal people who are actually subject to law).
People don't go to jail for errors like these.
Normal folks get fired for "mistakes" like this. Oligarchs don't.
Pretty big difference between getting fired and getting jail time, ain't it?
Still, I doubt anyone is getting fired over this. And the document was not written by an oligarch, lol.
And the document was not written by an oligarch, lol.
No, only about one, in order to pump up his career, and AIUI signed by him. "lol hahaha"
I replied to a comment about normal people being fired over mistakes like these.
Who is supposedly at risk of being fired if not the person who wrote the document?
Should Altman fire himself or what are you talking about?
Should Altman fire himself
He probably won't, so someone else should do it for him.
(Ideally, someone who seems so shady should of course never have ended up in such positions of power in the first place.)
What mistake? It’s an immaterial S1 for a SPAC. I don’t believe anyone including the advisors filing this would get in trouble.
That mistake is that it's not true. Altman has earned a name for himself in the truth-telling/transparency department. He wasn't briefly sacked for his fashion sense.
In the article they quote Marc Fagel, who previously served as the regional director for the SEC's office in San Francisco:
"Misstating the qualifications of executives is something the SEC takes seriously,"
Ok please remind me when the SEC takes action.
You'll probably be waiting until the cows come home for that reminder
People don't go to jail for errors like these.
Yeah, probably not. The problem is that they should.
The other problem is that people say stuff like "People don't go to jail for errors like these" as if that weren't a problem.
Your distorted view of Altman is probably causing you to go down the wrong path of securities law.
1) I don't believe even if malicious this would be a criminal case that leads to jail time.
2) The s-1 filings are prepared and audited by third party companies.
3) Most likely in this case, these positions are listed to outline any potential conflicts of interest. I am not well read on the matter but I could see that you would list potential engagements to note the conflict of interest even if they are not finalized.
The s-1 filings are prepared and audited by third party companies.
And signed by…whom?
That’s not really the point. The point is that this disclosure is immaterial to the spac and might have been true or in process to being true at the time of filing. Nothing will come of this other than a handful of people getting upset.
"in process to being true"
That's Altman-esque language.
Huh? It’s an S1 for a SPAC. Practically the whole S1 is immaterial. It was a disclosures page. Stop making something about nothing.
2) The s-1 filings are prepared and audited by third party companies.
does it completely release liability?
No but the point being that multiple people looked at this and most likely this is immaterial to the S1 and would never be on the radar for the SEC.
Can they? I thought SEC violations couldn't result in jail time.
The claim was repeated on the website of the SPAC itself until the story ran. And the SPAC is run by… Sam Altman.
And the SPAC is run by… Sam Altman.
Yes, but this comes with the assumption that Sam writes and reviews the website copy, PR statements and S1 form submissions, which any organization of significant size, the CEO does not.
I could believe Elon does though.
You absolutely read the S1 statement. What else would you be reading?
You would be reading your CFO or lawyer's summary of the important points. In this case, the only important point is probably literally just "We filed the S-1, everything went normally".
Do you use an accountant to file your taxes? Do you read the entire tax document they submit, or do you just look at their summary of "after this much in deductions, you owe this much"?
If you do read the document in full, good on you, but I assure you most people don't, in the same way that most CEOs don't read every single document the company's lawyer produces.
Yes I do in fact read those things, every VP or above that I know does too. A meeting covers high points that doesn’t mean you don’t have to do your homework.
Trust but verify.
Why do I get the feeling that if it said something bad they would have fixed it before somebody wrote an article about it...
And if they can't be trusted to get their own team biographies right, what can they get right?
If a run-of-the mill employee were lying about their experience, they'd be fired. But it's okay for the CEO because he's just so busy? What a double standard.
Did they use their own product to write the bio, and it "hallucinated" that he was chairman?
I saw the same odd excuse deployed for the aggressive NDAs he was embarrassed to find out happened
I think it's a mix of feels good to feel empathetic and say gee prolly just slipped by him, things slip by me, and then without interlocutors, could we ever have discussion? But it's an odd one
Its assumed that many many people involved in an already really cloaked investment vehicle know that Sam is not. Thats the point. Not that Sam wrote it, that the organization signed off on it.
It would be hilarious if some intern used ChatGPT to write the copy for the SPAC website and it hallucinated based off of being trained on the SEC filings.
I guess I don’t see the issue. The SPAC materials are largely pointless and immaterial until an acquisition is made. Purely a guess but I would assume they put these materials together when perhaps the board position was in play. Maybe he was fabricating it but it’s mostly a shrug moment for the SEC.
This is going to fall off the front page very quickly, just watch. The mods suppress anything that makes YC or anything they invest in look bad
Actually we moderate less, not more, when YC or a YC startup is the story. This is the #1 rule of HN moderation and has been since pg first explained it to me on the morning he started showing me how to moderate HN. If anyone wants to read about this, there are ten years' worth of past explanations here: https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu....
It looks like a mod downweighted this story but also saved it, along with your comment, for me to review. Since the rule is the rule, I'm reducing the downweight now. (Edit: I've since checked with the mod and indeed their intention was to follow the same rule - see https://news.ycombinator.com/item?id=40720026.)
This looks like a flimsy story from a couple of months ago about something from many years ago. It's hard to tell if there's any there-there but, given how this sort of story usually works, the odds seem low*. It's obviously getting attention not because of intellectual curiosity (the right reason on HN) but because of celebrity outrage (a popular human pastime but not the right reason on HN). It's also repetitive of many other outrage threads we've had in recent months—an additional reason to downweight it (beyond the outrage aspect). We never let articles like this stay on HN's frontpage normally—if we did, HN would consist of little else.
Still, I'm happy to reduce the downweight on this because that rule is key to the trust we have with the bulk of the community. I think most HN readers know that we moderate shallow outrage stories off the front page, and why; and I think most can appreciate that we bend that rule the other way for the sake of the above principle, even if they don't like it.
It doesn't stop people from making up false claims (edit: I mean about HN moderation, like the parent comment), but it does let us answer them in good conscience, and the rest is just standard internet dynamics (sinister speculation, etc.)
* Note the photo that describes Geoff as YC's "new president". That hasn't been the case for 5 years, and he stopped being president 18 months ago.
Hey Dang, I appreciate the transparency and general policy in such situations. As well as undowning the story
Do you think there will ever be in band signaling of when (or perhaps even why) a post was artificially depressed by moderation? Particularly aimed at cases where people are already using tools like https://news.social-protocols.org/stats?id=40710417 on posts like this where more people have already seen the story drop off and confirmation that it was via moderation than will probably ever see this comment in its later revival explaining how or why. Something like a "downed" with hover alt text next to the flag tag even. Apart from being immediately clearer to everyone without waiting for a comment/email that helps solidify against "and then 5 hours later they back tracked it slightly but not all the way with some other excuse why after being called out" type thinking/arguments. I mean you'll never get rid of that completely but it's a lot more powerful that the message existed there at T=30 minutes than T=5 hours.
Would typically email but this seemed like something good for an open message :).
I understand the theory—surely publishing more data would reassure everyone and help settle things—but this is based on an assumption which in my experience is false: that these dynamics are rational. I don't think that's how they work at all, so I don't have faith that earnestly publishing a moderation log would improve things. On the contrary, I find it easier to imagine it making things worse. Why? Because there would still be all the same suspicions and resentments, and now they'd have more material to work with.
I'm tired and doubtful that I've expressed this very well, but I've posted quite a bit about this over the years, so if the present comment doesn't help, perhaps https://news.ycombinator.com/item?id=39234189 will, or one of the other comments linked back from there.
Reading through the linked comment (and it's linked algolia feed) I'll add that public mod actions are less aimed at the making interactions with small and loud subset who will never engage rationally themselves any better and more aimed at helping the rest of the community easily ignore said subset. If the typical rational user will have seen many posts about YC critical news not related to celebrity gossip they will immediately know such claims are unfounded and have public references to it. Right now it's instead much easier to find 10 examples of this type of public comment thread where things may look gray or look at the public data analytics and find 10 examples that appear to confirm the viewpoint that topic $x articles are silently downweighted more often than they should be than it is for anyone but the mod team or those who send constant emails to provide even one bit of pre-existing hard proof to the contrary that doesn't carry the risk of looking like it was formed after the fact.
Specifically on a full moderation log your links also show how it would be troublesome in some ways I perhaps hadn't thought as deeply about. After all even if you moderate 9999 things right the same crowd is going to want to parade the 1 - now with more details as to why. At the same time I still think the same set of people are going to complain either way and the data isn't really helpful to them as much as the rest of the people anyways. I realize I'm extraordinarily unlikely to budge your views on this due to volume of experience/thought/interaction with this topic you've had but just count me in the selfish "I'd rather the log/actions were public for us average user and then your half a day of emailing was explaining the details to the troublesome bunch rather than with the average user spending time trying to understand".
I'll close with a smaller push of "and even if the site never does a full moderation log, not signaling moderated post/comment actions at any level only feeds that small group on believing everything unpopular they say is being censored 100% of the time instead of x% of the time". With that I'll leave you some peace on the matter :). Thanks for the time and reading through (yet another) user's view on this and for the good moderation of the site as always.
I appreciate your reading those past explanations (you're an existence proof that people sometimes do!) and I appreciate your thoughts. They aren't unreasonable, and it's not obvious what the optimum is, so I'm also just guessing.
I agree with you that the important thing is not to satisfy the litigious sort of internet user—the ones who take every explanation as fodder for fresh objections. It would be nice, of course, if they would take yes for an answer! and on rare occasions I've managed to pull it off, but it's hard, and ultimately too high a bar.
The important thing to satisfy the "rest of the community", as you put it. But the rest of the community seems to me already satisfied. Adding a new system like a moderation log wouldn't add to their satisfaction and might even diminish it.
I don't mean this in a complacent way! It has not always been the case that the "rest of the community" was more or less satisfied, and it took years to figure out how to get there. But we did figure it out (more or less), and the approach has been stable for a long time now. It boils down to this: we answer questions when people ask them, take their concerns seriously, and acknowledge and correct mistakes when we screw up. Because we do so much of this (I spend the bulk of my time doing it, many days), regular users eventually see repeated cases of it, and this lets them know that if they ever have a concern, a question, or just a curiosity, they can get an answer too.
This turned out to be by far the biggest factor in getting the HN system (community + mods + software) to be more self-regulating. What was lacking before that (mainly because pg didn't have time) was responsive, interactive moderation. It's important to understand that this is not a technical solution (like a moderation log). What makes things work better is human interaction, there's no substitute for that, and we're already doing quite a lot of it.
If the typical rational user will have seen many posts about YC critical news not related to celebrity gossip they will immediately know such claims are unfounded and have public references to it.
No, if it looks to them like many/most/all of those posts have been more or less subtly de-emphasized, the impression they'll get is the exact opposite of what you postulate.
Yeah, as much as the transparency is very much appreciated here, "yes, a mod totally did try to subtly bury this story exactly as you predicted and despite our official mod policies" sounds pretty bad, even if you undid it. Some visible "[Moderator Downvoting]" warning would help earn trust that mods are not in fact doing this to YC-related stories.
I checked with the mod and they weren't trying to bury the story but rather to downweight it less than we normally would, in keeping with the #1 rule I described upthread (https://news.ycombinator.com/item?id=40712900).
The intention was to downweight it enough that it wouldn't be near the top of the front page but not so much that it would leave the front page. That's the same thing I would have done, btw.
It looks like what happened was that more user flags came in and that, together with the moderation downweight, moved the story off the front page. Eventually we noticed this and restored it to the lower part of the front page, which is the standard practice for stories of this kind, which we'd never allow to be on HN's front page if they were about anybody else. (Though, as someone pointed out to me, this story wasn't negative about YC.)
It's hard to tell if there's any there-there but, given how this sort of story usually works, the odds seem low.
Not at all. The source links the relevant SEC filing.
https://www.sec.gov/Archives/edgar/data/1849056/000110465921...
On page 111 the false claim is plainly visible:
“ Prior to joining OpenAI, Mr. Altman served as President of Y Combinator from February 2014 until March 2019. Mr. Altman currently serves as the Chairman of the board of directors of Y Combinator.”
There is also a screenshot in the Business Times story showing the false claim was visible on the AltC website in April. You can go to the AltC site and confirm that Sam is CEO.
The story also includes comment from YC confirming that Sam was never chairman after it was announced he would be.
Dang, what is the basis for your comments on this story? The San Francisco Business Times has been a reliable and fairly pro business local source for years. Hardly tabloid journalism. Can you specify and explain one of the “false claims” you allege?
Oh, I meant false claims about HN moderation, like the GP comment: https://news.ycombinator.com/item?id=40711123 ("The mods suppress anything that makes YC or anything they invest in look bad"). That wasn't just completely made up, it's the exact opposite of the truth.
My comment originally included that explicitly, but I was trying to edit it down a bit and thought this would be clear from the context. Since it wasn't, I've added an edit there now.
it's the exact opposite of the truth.
One can hardly call it the exact opposite of the truth if the feared downweighting of this post actually happened.
It's commendable that it was corrected and that you are attempting to be transparent. But, it sure creates the appearance of a self-protective attempt to moderate.
Yes, I take your point and would have preferred the original downweight to be less for that reason. (Edit: I checked with the mod and originally it was - see https://news.ycombinator.com/item?id=40720026. It looks like user flags made the difference.)
Different mods inevitably make different calls sometimes, though. That doesn't change the fundamental practice, which has been stable for a long time (long before I took over HN long ago).
One thing I often point out in these situations is that "moderate less" (which is what the rule calls for) is not the same thing as "not moderating at all". Most probably the mod in this case was thinking along the lines of "I'm still moderating less in this case than we normally would, but I'll check with dang just in case".
You called the story “flimsy.” How so?
Crisis of trust in Altman reflects back on Y Combinator. They have significant financial motivation to downplay his behavior regardless of any moderators claims.
This looks like a flimsy story from a couple of months ago about something from many years ago.
"Here's me obediently following the rule (Oh, the world is so full of silly rules!), but let me editorialise a bit on why any story about how Sam Altman isn't the greatest thing since sliced bread should be disregarded..."
And you don't see how that undermines any "trust we have with the bulk of the community"?
I don't, because it doesn't. If the bulk of the community distrusted us, that would be by far the dominant issue on HN. That's how internet dynamics work.
The reason I post such long explanations when a case like this comes up is not because I think it will satisfy the disgruntled, chronically distrustful fraction of users (though that would be nice!) but because I know that explaining everything that happened, and why, is enough to reassure the bulk of the community.
(I wrote a bit more about this here if anyone cares: https://news.ycombinator.com/item?id=40720410)
I don't, because it doesn't.
It rather obviously does, AFAICS -- would we be having this discussion if it didn't? ;-)
The reason I post such long
The issue for me certainly isn't the length, but the tone (or "whiff" :-) of your comments.
explanations when a case like this comes up is not because I think it will satisfy the disgruntled, chronically distrustful fraction of users (though that would be nice!)
Hope I don't count among those. I've been a bit vocally critical recently, but that's just (at least intended as) pointing out where you slip up in your stated goals and policies, which on the whole I'm fairly certain you (all) are doing a pretty good job at.
but because I know that explaining everything that happened, and why, is enough to reassure the bulk of the community.
OK, so I'm apparently not in that bulk. Hope there's some "moderately sceptical" group between that bulk and your "disgruntled, chronically distrustful fraction" above for me to be counted in.
I dunno if it’s outrage-worthy, but the silent evolution of the YC blog post on Sam’s departure was weird, and nobody involved—particularly pg in his recent comments meant to be definitive—has managed to explain why that went down the way it did.
It's worth noting that `dang` is one of the people who noted the evolution happening.
You predicted it. Already on page 3. Algorithmic or not, we'll never know.
We answer questions about this kind of thing all the time (literally many times a day), both on the site and via email (hn@ycombinator.com). There's no question like this you can't find out the answer to. Whether you want the answer, or like the answer, or choose to believe the answer, are other questions.
The fact that there can be a thumb on the scale is (or should be) well known by now. Since the thumb is invisible, though, we can never truly know unless it's directly addressed, which was the case with this one. +1 to the other commenter's suggestion of visibly marking manual moderation activity (vs. algorithmic voodoo which is more understandably opaque). We've already got the [flagged] tag that transparently shows end-user moderation activity.
We will never be able to prove, but we DO know...
The post was moderated within 10 minutes of being submitted: https://news.social-protocols.org/stats?id=40710417
This tool is amazing. Thank you for sharing it!
Those were user flags. (I just checked. A mod didn't do anything until 38 minutes later.)
This is really dumb, and I'm not even a fan of sama, I think he's a snake.
Sam Altman was CEO and President. The fact he was not Chairman is not material. There's a thousand different explanations how that misstatement could have occurred. Maybe someone wrote that up, and sama didn't review every single line, and other people just assumed it was right. But it's not material. Being being the CEO and President of YC is much, much more important and more material than Chairman of the board, which is largely a figurehead position.
No one is going to care and this is extreme nitpicking. They should fix the misstatement, but to claim some sort of conspiracy around this is stupid.
But it's not material.
the corporation President, VP, Secretary, Treasurer, and Board of Directors and Chairman are legally defined titles, they are material titles, they imply fiduciary duty, loyalty, etc. (CEO isn't really a thing, it was made up to point out whether the President or Chairman wielded the ultimate power in situations where the largest shareholder might be more active than passive)
Materiality depend on the question at hand.
SEC filings would be enough for materiality one would assume.
For what question? What is it evidence of exactly?
E.g. It doesn’t tell you what sama had for breakfast.
Not an S1 and especially for a SPAC. Nothing is truly material until an acquisition is made.
You would not be correct. If It has to influence something. Being chairman of the board is much less prestigious or important than CEO or president which is actually doing work. Chairman of a board is a figurehead position.
As I read the statement in the article, he didn't claim he was the chairman of Y Combinator at the time he was actually President. He claimed was the chairman at the time of the filing, 2021:
"Mr. Altman is an entrepreneurial and thought leader with a proven track record of growth investing within the technology industry. He is also the co-founder and CEO of OpenAI, a leading AI research and deployment company, and the chairman of Y Combinator"
And as mentioned, "president" and "chairman" aren't fuzzy questions for the law. Saying you're one when you are the other is a lying, not "fudging a bit". Lying in generally, say at cocktail party, isn't against the law. But lying to the SEC is definitely a thing that can have penalties.
And, yes, the article takes an appropriate "why would you even do that?" tone since it seems unlikely he'd get some financial or other payoff for such behavior. The main impression it gives me is of a pathological level of hubris/ego. Like a manager who j-walks in front of cop on company time.
Lying is a very difficult burden to prove and requires meeting several criteria:
1. Intent to deceive, so it can't just be that the statement is false but that it was intentionally false as opposed to what is likely a mistake made by a third party. As was noted elsewhere, Sam was positioned to be Chairman of Y Combinator after stepping down as President, but that never happened. It's entirely possible that whoever wrote this did not become aware of that change in circumstances.
2. Reliance and harm. The false statement must be something that was or can be relied upon by potential investors that could cause them harm. It's possible an investor can rely on this fact but it would be very hard to demonstrate harm from it.
3. Materiality. The statement must be significant in that the information could sway a potential investor towards the investment. I'd say misstating credentials is material and so this element does count.
Taken altogether however, this does seem like a whole lot of nothing.
100% what I was going to say.
This is a just a misstatement. There's plenty of valid things to critique sama on, but this isn't one of them.
It can have penalties but most like not for a disclosures bio page of a spac.
It’s a misstatement and most probably an oversight at worst. It’s only material if it affects people’s decisions in investing. Theres no material amount of prestige associated with a board position over an actual working position like ceo or president.
If I claim in a job application that I have worked several years for, let's say, Google, and it later turns out that that was a "mistake" (I was only promised a job, but never got it), I could definitely get fired. Why should Altman expect to face no consequences whatsoever from such a "misstatement"?
Not really a fair comparison... how about if you _did_ work for Google, your role was Principal Engineer on Google Buzz and you were the tech lead for the team working on it. On the Google Buzz landing page it says rob74, Director of Google Buzz, led the project
And btw, there was noone with a Director title working on the project, you were in fact the main project lead
I think these types of posts are extremely nitpicky and not useful. From my experience, these sort of SEC filings are compiled by associates in a biglaw firm under a lot of time pressure, with file names like "version 41.B FINAL EDIT". Things get lost in translation. The signing off parties on the document are not sama: https://www.sec.gov/Archives/edgar/data/1849056/000110465921...
I think someone made a mistake. Also, from looking at the sig page the signing parties are others...
I had the opportunity to chat with Sam, and I would say that the majority of publicly written opinion about him is false.
He’s a pretty genuine guy with a lot of advice and wisdom. He definitely doesn’t come off as a capitalist but more of a helper type — someone who helps people become whole.
I felt like OpenAI and the sector in general made a lot of sense — now he can help everyone in the world.
As for the issue at hand, for one it’s doubtful he prepared this, but moreover, I have seen quite a many resumes in my day and not one wasn’t padded so to speak.
You are very perspicacious if you managed to figure that out after a single chat.
Chats + Many emails
He always helped when I ping'd him. ¯\_(ツ)_/¯
Thank you for everything Sam.
It’s his job to build relationships and trade favors. I don’t see how it contradicts the majority of publicly written opinion about him.
To be fair, Lord Rasengan is a sovereign head of the Joseon Empire cybernation according to their own bio, so Sam was just conducting international relations for openAI's next country to exploit. Be careful you don't anger Lord Rasengan; they could throw you in internet jail.
He thinks Worldcoin is a great idea.
It's a bad execution for a wholesome, world helping, idea.
It’s a bad solution for the wrong problem
> He definitely doesn’t come off as a capitalist but more of a helper type
A capitalist is an owner of the means of production. It's as simple as that. I think Altman qualifies, especially to the African workers who make $1.50 an hour to get PTSD[1][2][3] in his AI sweatshops.
[1] https://www.ft.com/content/ef42e78f-e578-450b-9e43-36fbd1e20...
[2] https://time.com/6275995/chatgpt-facebook-african-workers-un...
[3] https://www.theverge.com/features/23764584/ai-artificial-int...
How does a mistake like that get made 3 or 4 years later when he was never the chair, and the only official source for that had been edited within days to remove that, and apparently there has never been a YC chair at all?
I imagine the associates googled for it and repeated whichever source existed at that time.
The media obsession with Sam Altman reminds me of the old dutch saying 'high trees capture a lot of wind' (hoge bomen vangen veel wind)
I suppose that's why many HNWI individuals consciously stay below the radar.
In Aus and NZ this is known as "cutting down the tall poppy": https://en.wikipedia.org/wiki/Tall_poppy_syndrome
In Japan it goes, "The nail that sticks out gets hammered."
(Well not a metal "nail", but a wooden post.)
In my section of America, it's "the loose board gets nailed", which is just enough of an innuendo to make it stick.
Hoë bome vang die meeste wind (Afrikaans)
It would either be "Die hoogste bome vang die meeste wind" or "Hoë bome vang veel wind", if my idioms serve me correctly
The problem is when your wealth is inseparable from your celebrity. Altman is the face of AI. If he kept his foot out of his mouth he wouldn't be rich.
In Chinese there is 树(tree)大(big)招(attract)风(wind) but it is used in a way that implies the tree is innocent and the wind is at fault.
Definitely. Though part of your job as a leader typically is also to catch that attention and direct it to serve the organisation.
Unless you're a high net worth HNWI individual, right?
'high trees capture a lot of wind'
Makes me feel better about being short (on a good day I'm 5"3')
The real problem is that Altman has legitimately destroyed public trust. We already have way too many immature robber barons playing the oligarchy game for the likes of such as him to be positive for society in the end. If he had a conscience, it appears that he's sold it already.
The real problem is that Altman has legitimately destroyed public trust.
This might seem true in our technical buble, but most of the public probably has no idea who he is. They try ChatGPT for a work task and think it's amazing and love it.
I've seen this happen many times, and once one person in a small team/company tries it, everyone else starts using it and they also love it. These people don't know what OpenAI or who Sam Altman is and there hasn't been any trust broken because ChatGPT works for them.
Apple just gave the company under his leadership a huge vote of confidence.
And? That's the company and not him as far as what iDevice users will care about. Again, even if they bother to learn the name of the company that makes chatgpt. Hell, they might even think that is the name of the company and not the product. I don't really understand your point
That's what they thought with Qualcomm.
i don't get the reference
The upstream comment I was replying to mentioned that SAMA destroyed public trust. If this were the case, then what he had done might be enough for a massive company like Apple to hesitate on making such a deal. And there are alternative routes Apple could have taken (maybe not as good as GPT4, but close in a fast moving space.) It's true that users may not know who the company is (let alone know about SAMA) - but he does have outsized control over a company which "isn't him." He had the power to reverse his sacking, make changes to the board, and is now considering changing the structure such that OpenAI is no longer a non-profit entity.
Where there's smoke there's fire. You don't get to a point where you "destroy public trust" without also destroying investor confidence. The upstream comment mentioned "public trust" - but there were also trust issues with the board. These issues then put further investment in limbo until he was able to change the board.
My point is, that any trust issues aren't likely what they seem given that Apple made such a large bet with the company. Sure, it's "only" a distribution deal. But Apple needs to know that OpenAI will continue to have access to the resources they need to continue servicing massive markets. The trust may be wobbly, but not destroyed.
But Apple's deal with OpenAI is superficial at best. It's not their entire approach to AI. You can use Apple Intelligence and not once submit something to ChatGPT. To me, that means that they are only using OpenAI as a crutch until they've improved to make it no longer necessary and/or gives them an option to swap out which ever 3rd party model is the new hotness. It's less embedded than having Google be the default search engine as ChatGPT isn't the default AI model and is absolutely opt-in with a warning message stronger than any data sharing message from anything they've offered.
Apple is pragmatic; they know ChatGPT has the mindshare of the public (90% of whom have never heard of Claude, etc.). But make no mistake, they will switch to their own similar offering once it's up to par.
Reading this comment section, I'm realising that smart people in tech are still just people. They love to gossip, shit-talk about others, prop up cartoon villains to hate on.
I'm a big fan of sama. It's been said he's a snake. I hope he is. And he will do things you will hate and be successful doing it. And in doing so, will teach you a priceless lesson. DGAF about other people. Be yourself, look out for yourself and fucking bring it and step on the stage, as you are.
You don't have to be a perfect and excruciatingly boring paragon of virtue before you become a public figure. How wonderful!! It's ok to be a character. As long as it's your character.
It’s even worse on here since the HN audience seems to worship Altman, although table seem to have turned and he is a bad guy now. Can’t make this stuff up
The love and hate comes with the territory when you're a public figure. I don't think he needs defending, and I don't think there is any lesson to be learned if a famous person you dislike is successful.
You put together perfectly. Haters don't realize sama actually put together something tangible and believable while others either berated looked down upon each other or fried their brain theorizing transformers.
That level of contempt for the people around you is exactly why tech companies can't even create something as boring as a taxi company or corporate shuttle without raising the ire of everyone else.
It's not Ok to be an asshole or a snake. You calling that "character" doesn't help; assholes and snakes are not OK characters to be.
(And I think in order not to be able to grasp that, one would have to be an asshole or a snake oneself.)
I've never met him in person and I don't usually judge a book by its cover, but he gives me a creepy vibe. It is difficult to put in words because this the first time I feel like this about someone I've never met. Something feels always off when I look at his photos or videos. This has been happening long before all this OpenAi drama.
This may give some insight on the creepiness you sense, it involves DARPA https://www.newyorker.com/magazine/2016/10/10/sam-altmans-ma...
The idea that "DARPA is incompetent" because they promote hair-brained ideas, is one of their biggest ruses. Of course they're competent.
Corollary: Elon Musk https://www.reddit.com/r/ChatGPT/comments/1dgyujd/comment/l8...
The idea that "DARPA is incompetent" because they promote hair-brained ideas
Hare-brained. It refers not to hair on the brain, but to hares not being the world's smartest animals.
Yup. Personally I think he might have dirt on Paul Graham.
I'm not a Sam Altman fan but this really seems like nothing. If he was personally going around telling people he was YC's chairman that would be quite different, but this is just a few documents that he obviously didn't write himself.
Lying in an SEC filing is not "just a few documents".
You don't think people make clerical errors in SEC filings? What was Sam supposed to have gained by this?
The document was signed by Michael Klein and Jay Taragin (whoever they are), not Sam. But that wouldn't make much of a story, so Sam it is.
I think you're completely wrong. It's worse than going around saying stuff, it's publishing that statement on official documents for multiple years.
From the article:
Yet, Altman has been claiming the chairman title for at least three years in securities filings and a website associated with the AltC Acquisition Corp., a special purpose acquisition corporation, or SPAC, that he registered in 2021.
Why does the guy claim to be a chair? It's what people place their asses and fart on. Is it some kind of weird fetish? Hey, I'm not judging, just asking the important questions.
He would still be leading board meetings even without a board vote perhaps?
When you or I lie on a resume the background check knocks us out of the job. But when the CEO lies on a resume, nothing happens.
I think it’s abundantly clear that if OpenAI fails, powerful people will face prosecution. It’s an uncomfortable but unavoidable part of our justice system that we don’t like to prosecute people until they’re no longer winners.
It's far from clear.
Dozens of finance executives wiped a decade worth of economic growth off the United States the year before I graduated high school. My generation will never recover. They left their jobs with more money that I'll make in an entire lifetime. None went to prison.
If someone in your generation (graduating high school 2009) obtained in demand credentials or skills, then they should have reaped enormous returns from 2013 to 2024.
The only people who really “lost” in 2008 were those who were overleveraged from going into debt for houses or education, and then had to default due to income loss in 2008 to 2011 or so. Or those who panic sold all their equities.
How is a new college grad supposed to reap "enormous returns" without significant savings?
From entering a hot labor market with rising wages and investing?
The alternative is entering a cool labor market with stagnant or decreasing wages.
Obviously, people who already had assets made out better, but that is not a possibility for 99% of high school/college graduates, so useless to consider.
I think people really over-estimate the wage growth differences between good and bad labor markets.
Median usual weekly earnings are (in 1982 Dollars) Q1 2024: 365 Q1 2013: 331
So over 9 years, including the largest bump in wages in generation during the pandemic, wages grew by 1%
From the early 80s until now, wage growth was like 0.4%
https://fred.stlouisfed.org/graph/?g=1pagn
I restricted my claim to those who:
From plumbers to programmers, there probably was not a better time in history to get paid. That is the nature of a decade of 0% interest rates, unprecedented in US history.
If you did not get paid, it was because you were not shopping around for a buyer willing to pay more (not ascribing blame).
IMO nobody highly typical reaped major returns from '2013 - 2019' unless they already had excess income.
Even restricting to CS (and I haven't tracked my graduating CS classes that closely from the '08 - '11 range) my memory is that many of the smartest ones didn't end up working in CS for reasons that were never clear to me and sort of muddled by for years. (To a certain extent I'm still not sure how college grads who didn't get internships or work in the field are supposed to land their first job in the field if they're great at c and algorithms but can't and won't market themselves or move.) The only ones who made out like bandits were the ones who landed internships at places like Google or MS or had preexisting connections or niche interests of some form. A solid chunk ended up at big old companies, like IBM, Lockheed Martin, etc. They mostly made out well because getting paid $60k - $90k out of college around then made it easy to buy a house, and because they worked somewhere that made a job switch or career progression work out well enough to start getting paid more circa mid '10s. That said, a solid chunk ended up in smaller companies, government, or non-tech companies that needed devs, and I think for many of them it didn't work out or lead to years of low pay before being able to get more on track. For many, it seems like things didn't really lead to big benefits until the pandemic, which I think unlocked opportunities for more folks to work remote.
Back to speaking more generally, I think many more people probably could have reaped the benefits of home ownership than did so in the '08 to '19 interval because houses seemed to stay cheap all over the country for much of that period. But many folks in the age bracket seem like they either (a) didn't realize they could afford to when houses were quite cheap, (b) actually couldn't (student loans?) or (c) were afraid to, because that whole generation never feels secure in a job or their ability to find a job if their current one cuts them, and got somewhat chilled by the foreclosures they saw in the '08 crash. Many of these folks I think only realized how this was screwing them when rents started to skyrocket, though I don't think I clearly have a sense of how student loans play out - I do think college enrollment as a percentage of high school graduates was pretty high, but I'm not sure how big a percentage of those enrolled in high cost vs low cost situations in both absolutes and loans.
But IMO it's important to realize that for most typical workers in those years, median income was not high enough for single income to invest significantly beyond their house if they did save towards home ownership. The number of folks who didn't know how to find anything gainful to do with their degrees and so went back to school for master's or higher in those '08 - '11 years seems significant to me. I personally know plenty of folks who got masters degrees or better in various fields and basically eke out a living doing stuff like writing website copy, teaching at charter schools, and various other things that don't pay well and seem likely to be sort of demoralizing for someone with an advanced degree.
So IMO it's only with two income households and above median incomes that you can realistically do both (invest and own housing) over those years, particularly if there are any student loans. And during those years marriage ages have gone up.
The thing I'm seeing now is that folks who did get housing are kind of stuck doubling or tripling their payment to get a large place if they want to have kids, etc. Which will likely further negate many folks' ability to participate in the market. Or lead to further birth rate decline.
Sure, it was only the executives and not the government bailing them....
Yep, Elizabeth Holmes was bulletproof until the cracks started to show through. Same with Lance Armstrong, and many others.
Was Holmes ever bulletproof? Anyone with scientific experience was calling BS on her from the beginning due to the reality of current technology and physics.
I remember only knowing her because there were posts on this website calling her out very early on.
Yes Lizzie was absolutely untouchable and invincible until the façade came crashing down.
Humans Want To Believe.
https://m.media-amazon.com/images/I/41x244sdEXL._AC_UF894,10...
I'm not familiar with the whole situation - what would they be prosecuted for? Violation of copyright? Or is this some sort of 'everything is securities fraud' situation?
this can be considered as essential job function, so checking the quality of lying is a part of interviewing process.
I worked at a ~$20M food company where we had a new CEO coming in. Management/Board were over the moon, saying "This guy has been running a multi-billion $$ company for years."
Curious as to why someone would transition so abruptly from the major leagues to the minors, I searched on his name and the first hits were about an active SEC investigation into him for witholding payments to suppliers (small and medium farmers) to inflate the balance sheets to give it a higher valuation for a buyout. He actually turned out to be more effective than the previous 2 CEOs and I forged a perfectly fine working relationship with him, but it made me aware that business people will ignore any variety of moral/legal issues when the dollar signs are in their eyes.
I believe he put the SEC thing behind him, avoiding jail time by not quite admitting nor denying guilt and writing a ~$2M personal check.
Just the sort of personality type that we need leading the most potentially dangerous new technology in the past 50 years.
While I get the sentiment, things have a way of working out. Maybe he is actually the best person to do it...no one knows who the best person for an undefined situation will be.
There is no "plot armor" in the real world, protecting humans from having bad things happening
Things do not, in fact, have a way of "working out"
If we don't want bad things to happen, we have to actively stop bad things from happening
Depends on how rich you are.
Problem is, the "bad things" the rich want to stop from happening are usually what the rest of us are doing to stop bad things from happening...
Who do you think should run OpenAI then?
The definition of survivorship bias because you literally are alive to say this.
Tens of millions have been killed by CEOs of tobacco companies, and that's just one industry. Energy companies will likely cause the death of billions.
We got rid of many dangerous American oligarchs in the Gilded Age by using govt to break them up. Things don't "just work out," people sacrifice huge amounts of time and energy to fix them.
didn‘t the “tens of millions” of humans who died have any agency? And by definition, aren’t “energy companies” helping fix the problem of “energy poverty” which increases human welfare?
https://en.m.wikipedia.org/wiki/Energy_poverty
In both cases, the agency of individuals was undermined by regulatory capture and the publishing of false studies (despite having real studies that the companies buried).
For decades, people didn't even know it was dangerous to smoke or have lead in gasoline or burn fossil fuels because the information was systematically suppressed by the producers of those products.
Is the best person for any position (especially one with tremendous implications for the future of humanity) someone with his character?
Does it matter if the person is the best or if people think they are? Even our government is led by people that think the person is the best for the role even if they really are not. Why would anyone care about this particular company when that's their view on the gov't?
I would submit that if we care about outcomes then it matters if someone is the best suited for the role.
There will always be subjectivity, and “the best” is going to be at least partly defined in terms of who is available. Nevertheless we could insist upon certain standards. Not being a pathological liar, a sociopath, deeply conflicted or self-interested, and so on, seem like good starting positions.
I’m not arguing that we live in a reality that reflects this, of course. Just that it would be nice to.
Yeah, well, you know, that's just, like, your opinion, man.
The only potential danger of said technology is that it may be confused as something trusted, but as shown above, trust is a fundamentally flawed concept. Which means that if danger actually arises, it was self-inflicted.
This is startup land. We call it a Painted Door test of the truth. He's just selling product roadmap before things are feature complete, we all do it.
Please don’t say all… I for one have never done this.
My comment was meant to be satire.
On Reddit he doesn't even bother lying, he just straight up admits being a conman.
Yishan Wong describes the con: https://reddit.com/r/AskReddit/comments/3cs78i/whats_the_bes... Altman admits it's true: https://reddit.com/r/AskReddit/comments/3cs78i/whats_the_bes...
To my eyes, Altman appears to be joking.
Altman saving Reddit makes me like him more, not less. Imagine rooting for Condé Nast letting Reddit rot to death, Yahoo acquisition style.
He used Chat GPT to hallucinate his CV, I guess.
If Elon Musk were still alive, he'd be soooo jealous.
At this point, I hope this is well known.