I wish he was saying goodbye.
I know that Sam Altman is rich, is controversial, is in charge of OpenAI, yadda yadda yadda, but I was wondering what his expertise was, what he actually does to earn attention, because it’s not as if there’s anything I can point to and say, “Sam Altman was responsible for that.” He just seems to have a lot of money that he spends on other people who do a lot of different things. The Washington Post ran a long and laudatory article that I read to find the answer.
Here’s the shortest summary I could find.
In a Silicon Valley milieu in which shooting star companies often give birth to cults of personality around firms’ founders, Altman has stood out. An investor with a dizzying array of interests, Altman might lack the singular focus of a Steve Jobs — or the sophisticated technical skills to create the products he sells — but according to fans and rivals, he has had since an early age an uncanny entrepreneurial energy and a force of will that inspires others to do their creative best.
Jesus. That’s a particularly vacuous bit of empty PR, isn’t it? So he has an “uncanny entrepreneurial energy” and “force of will,” and what he does with that is inspire other people to do the work he can’t.
Reading between the lines in the rest of the article, it becomes clear that what he has is lots of money, which he acquired by becoming best buddies with people like Peter Thiel, and that lots of people want to say nice things about him in hopes that he’ll give them some of his money. He’s a college dropout with no particular skills, other than money.
The article ends with this bullshit.
“He’s the kind of founder that can bend reality,” said Hemant Taneja, a friend of Altman’s and managing director of the venture capital firm General Catalyst, adding that Altman had invited him to invest in OpenAI but that he declined because he couldn’t understand the company’s complex structure. “By creating the fastest and most popular consumer application of generative AI, he showed us the art of the possible. … This is the first technology where every CEO of every company in every industry is now thinking about how to do AI in their business. He made that happen.”
So he’s a hype machine, aided by highly placed friends, who promotes the AI buzzword which people are already beginning to back away from. OK, got it.
One of our modern problems is that the structure of our society incessantly pushes money-shuffling assholes to the top of everything, rather than competent people with actual constructive skills. I’m not impressed by Altman or others of his ilk…which is, unfortunately, why I’ll die poor someday.
You and a few billion other people on this planet will also die poor so you’re in good company.
Hopefully, the days of rich white men, (and their begging sycophants), celebrating themselves, will soon be over.
A quote I’ll remember.
In the 19th and the 20th centuries, some genuinely talented engineers built a fortune on inventions.
Today, the inventors are the first victims of the cleptocracy in the chain of tech development.
I honestly wonder if the end point of capitalism is a few trillionaires shuffling some money around between them as everything they need is automated and the rest of humanity has long since starved to death or drowned in the rising seas.
Got it: he’s a rich kid with waaaay too much time on his hands. I’d rather live and die poor (already accomplished the former) than be his buddy.
I look at history for examples, as none today are exemplars for excellence.
Tesla, knew how to build things, was quite proficient at it, scattered interests claimed most of his fortune, he died intestate. Edison, a peer and former employer of Tesla, knew how to build things, better at hiring people to innovate by assembly line.
Compared to Altman and Musk, can’t build jack shit, can pay people to build shit, great at selling snake oil, so they’re great inventors or something.
Yep, they’re great at inventing jack shit. And spending money.
Blundering into figuring out how to hire people who can innovate. Then, fucking things up. See Musk’s refusal to bolster his launch pad, which nearly destroyed his entire launch site and spewed dirt and rocks across an entire town and wildlife sanctuary. See Twaddle, Xit.
They both know how to sell their snake oil, AI or hyperloop, neither ready for prime time and generally, not really living up to the hype. With AI being, well, a joke currently (to judge by its output quality) and the hyperloop being the size of a small parking lot, not some intercity transport, his Boring company being exceptionally boring and not boring very far at all, Tesla having yet another recall for safety (something about ejecting occupants in a crash, not to mention the autopilot that isn’t).
Which points directly back to US culture and money is god. Totally infallible, totally ineffable. Despite a repeated track record of epic failures.
See Trump…
There are successes though with that trio, successes in bankruptcies. It was one of Trump’s only innovations, bankruptcy billed as a success.
Next week, stage 4 cancer billed as a successful survival strategy.
Don’t underrate people management as a genuine skill. ChatGPT is a truly revolutionary product, the result of hard work by some extremely intelligent tech people, as well as the orchestrating of the labor of a vast army of badly-paid human workers (see Kevin Drum’s recent post on the labor force behind AI training). To make this work, you need to hire the really smart people, plus several layers of competent and motivated managers, and keep the whole team focused on a goal. It would seem that Altman is able to recognize people who are smarter than he is, able to get them enthused so they’ll work for him, and to keep them enthused and pointed in a coherent direction. That’s not a common ability.
Yes, management is a skill. Note this other comment from the article:
He doesn’t do much managing.
When you are going to build wealth it is always good to have a foundation of wealth to build on.
I read this thread and I still don’t quite know what Sam Altman does.
OpenAI isn’t even a public company.
It is owned by Microsoft, one of the richest companies on the planet at 49%. The other 49% has to be a collection of rich venture capitalists.
The last thing OpenAI needs is money. They are surrounded by huge quantities of money.
So Sam Altman’s money isn’t needed at all.
Altman does have money though. He made it as a venture capitalist.
He is one of the founders of OpenAI in 2015.
“OpenAI was founded in 2015 by Altman, Elon Musk and others as a non-profit research lab.”
His money isn’t needed and he doesn’t run OpenAI.
It seems he was one of the founders of OpenAI, which isn’t trivial.
He also picked the people that actually run OpenAI, which also isn’t trivial, if they are very competent people who end up accomplishing a lot.
So, while Sam Altman can claim to have founded OpenAI and built it to its present leading position in AI, by now it isn’t at all clear what he is needed for.
He might be supplying critical strategic and tactical leadership or he might just be occupying a large office somewhere. I can’t tell from anything I’ve read.
Well, there is this anyway.
It’s no secret I really don’t like Elon Musk for a lot of good reasons.
He is a kooky neo-fascist who treats all the people around him badly, like things, not persons.
OpenAI might have missed Musk’s money, but they wouldn’t miss Elon Musk at all.
It could be better off as like the orange menace, every business he touches dies.
I remember when Facebook was still exploding. The venture capitalists I was talking to, back then, when I still talked to venture capitalists, were all saying “we want hackey sack playing genius founders.” Uh, not tech geniuses who broke open a whole new market? The problem with venture capital’s approach is that it favors funding what worked before – so they get a lot of rich white goofoff college dropouts – when they are looking for the next Bill Gates. Not that Gates was a great genius. He’s very smart, but mostly smart enough to listen to people around him who had different ideas of where computing should go, and had the ability to sort through all the options and come up with a set of options that worked pretty well.
All of this so much, which made it absolutely flabbergasting when Congress had Altman testify on AI a few months back. He just plays with other people’s money. He has nothing useful to say on the subject, and of course he said nothing useful. Gah!
Yeah, we’ve had waves of rich people. Sure, some people did smart things to get rich one or two hundred years ago. That’s really difficult right now because the moneyed interests are primarily interested in making sure that doesn’t happen. Because now the way to get rich is to be born into it or suck up to someone who was. That’ll go on for a bit longer but we can already see what the next wave will be like. It’ll be people like Sam Bankman Fried and Elizabeth Holmes. Fake it ’til you make it won’t just be the desperation play of the poor anymore, it’ll be how to defraud the stupid, entitled great great -etc. grandchildren of wealthy innovators from the 1800’s & 1900’s. Until all of those grandchildren are left with nothing but talk like Donald Trump.
It’s like he’s the real life version of that joke from The Justice League where Barry Allen asks Bruce Wayne what his superpower is and Wayne replies “I’m rich.” It’s the same with guys like Altman and Musk. In a world where there are people that do real, amazing things that might as well be superpowers (though ones earned by hard work, study, and talent), they’re the average guys whose only superpower is wealth.
There are two words for Altman and his ilk: Maxwell’s Daemon (or Demon, his handwriting was atrocious). One would wonder if those extolling the virtues of Altman — and Zuckerberg and so on — were aware that decades ago there was a mathematically-sound, clear demonstration that manipulating not pure heat but “just information”† also violates the second law of thermodynamics in a closed system. OK, maybe one won’t wonder when one realizes that the canonical demonstration was by a foreigner, not in English, with lots of math, for an audience of physicists and chemists with PhDs, and controlled in English through a really inept translation in a rather obscure journal having little to do with, well, physics or chemistry or even “information theory.”
So not only are Altman and his ilk not “scientists” (and managers of scientists need to be scientists, too, if not necessarily at the cutting edge themselves; after the time I spent in the DoD during the Counterreformation, and in dealing with lawyers and judges on scientific/related issues, I’ve seen the converse up close and personal) — they’re actively ignoring reality. Very much like, as Gary Larson noted all those years ago, the real reasons the dinosaurs became extinct. Sort of like the predictable dot-bomb implosion, which followed by a quarter of a century the unlimited-cheap-energy implosion, which followed by a couple decades the… oh, never mind.
† And, by necessary implication, “just money.”
Amen to that. This is exactly what has just happened in New Zealand, assisted by Conspiracy Theory Nut jobs.
Reminds me of an episode of Newsradio in which Bill (Phil Hartman) interviews a “business visionary” who, at the end, identifies his vision as…computers (though he doesn’t have one yet).
Red fucking flag right there. Anyone who pals around with Thiel is not a fan of democracy.
AI, it needs regulation because it’s daaaaaaaaaaannnnnnnnnnnggerrrrrroooooooooooouuuuuuuuuuussssss!!!
Then, one examines the current state of the art and finds a Norden bombsight.
All of the super secrecy around the Norden bombsight wasn’t from the government, but from the CEO of the company, ginning up all manner of obstacles for what, per a German evaluation of a captured bombsight, was about equal to what the Germans already had. So, the company president made it like it was a super raygun from hell, that absolutely had to be even more secret than the Manhattan Project, which was actually a supreme secret.
Invent a threat and super capability, guarantee sales in your market, especially if you can get the government to waste effort classifying and restricting it. Even if your super bombsight missed the target at Nagasaki by three miles.
In the video @ 14, Altman testified basically that users are smart enough to check. I’m aware of two lawyers who no longer are allowed to practice law that prove otherwise, after submitting ChatGPT arguments to a court.
Jaws @ 17, Counterreformation or the cultural revolution? ;)
I’ll just get my hat…
wzrd1 @21: Just call me Capt Torquemada, and hand me the soft cushions…
We could’ve used a little culture to rationalize a revolution, but there was more culture in the average 8oz cup of yogurt than in a typical political appointee (non-party specific).
None of the people on this thread, including the host would have been able to get the right people to create ChatGPT no matter how much money you had. You wouldn’t have even considered it a possibility worth spending any money on. That’s what he does.
Somehow, Sam Altman appeared and instantly a large body of scientists and philosophers and academics who studied things like large language models manifested themselves to create ChatGPT.
You’ve got the order backwards, guy. Altman is a salesman who showed up late to the party to monetize everyone else’s work.
This might be half true.
Most of us aren’t computer programmers or venture capitalists.
This is a biology oriented blog and PZ Myers is…a professor of Biology at UM Morris. I’m sure if you gave Sam Altman a microscope and a colony of spiders, he would be completely, totally lost and wander off to play video games or whatever he does for fun.
To be sure, a few of us are very definitely computer programmers of one sort or another and may well be able to do as much or more than Sam Altman, given the money and resources he has available to him by being a tech bro in Silicon Valley.
I think you’ve got you’ve got your second assertion without proof or data wrong, “no matter how much money you had.”
I’d bet a lot of us could run circles around Sam Altman, if we had access to the money he had and has.
This is all irrelevant anyway.
The comparison PZ Myers is making isn’t between Sam Altman and a somewhat random collection of commenters on the blog, Freethought Pharyngula.
The comparison is between Sam Altman and the much larger set of say, residents of North America.
Ed: “None of the people on this thread, including the host would have been able to get the right people to create ChatGPT no matter how much money you had.”
That wasn’t the question we are answering.
The question was, what does Sam Altman do?
I can and did give him credit for being a founder of OpenAI. That was in 2015.
It’s now 49% owned by Microsoft, one of the richest companies on the planet which means they don’t have any problems getting money or people. In case you aren’t clear on the concept, you recruit talent by…paying them lots of money.
So what does Sam Altman do now?
We still don’t know.
I’m guessing he has a big office that he rarely shows up for, and he plays lots of video games, hypes whatever needs hyping that day, and is looking for his…Next Big Thing, like Musk and the rest of them do.
I had to look it up.
OpenAI is 49% owned by Microsoft, a company that somehow knows a lot about computers and programming. It is their business and I’m right now staring into the screen of a Microsoft Windows computer.
Microsoft is the second largest company in the world with a market cap of $2.97 trillion. It is safe to say that OpenAI isn’t going to run out of money ever.
Excellent summary of what Sam Altman is good at in this piece- https://www.todayintabs.com/p/defective-accelerationism
TL;DR is not very much.
Making the connections that allow a venture to thrive is a skilled job with relatively few people actually good at it. It’s one I know I suck at.
Once those connections are made, yeah, that part of the job is done. Don’t worry about them falling into unemployment, there’s always more ventures to set up.
There certainly is a problem that the public face who gets the project set up gets all the credit in the public eye. It doesn’t mean the public face is useless, which is the opposite extreme you’ve swung to just to be contrary.
Other examples of this type of CEO to consider are Sam Bankman-Fried and Elizabeth Holmes. Both pretended to be tech-hip CEOs with ground-breaking capabilities and world-dominating goals. I do not think either of them actually understood what their companies did or how they did it, thereby calling the question of what they did. Manage? Well, maybe. Lead? Well, maybe. I submit that many “leaders” are more lucky than substantive. They surround themselves with people who can actually do the work, then they use all sorts of techniques to stay on top of the pile (e.g., bully, nag, call in authorities – read “Bad Blood” or “Number Go Up” to get blow-by-blow examples).
@24 Herding talent is a very nontrivial part of what good CEOs do. Leave a large collection of brilliant scientists to their own devices and witness how a wonderful amount of papers will emerge together with zero actual products that the masses can use.
@25 As someone who works at a company with A LOT of money to throw around, I can assure you that is not even close to enough for attracting top talent after you cross a certain threshold. Vision, reputation, and yes hype are often much more important than money to the talent. Witness e.g. Musk’s companies as an example of what kind of magic hype can achieve. Vision, leadership and hype are likely essential parts of what Sam Altman does, and clearly the employees value this a whole lot as evidenced by the recent events.
OK, here’s a fine parallel.
Who is the true father of the atomic bomb? The CEO, if you will.
If you say Oppenheimer, you’re wrong, as Oppenheimer worked for that specific boss, who isn’t given credit. Odd, given he was also responsible for overseeing the construction of an odd building that had four walls and a spare wall built in – the pentagon.
General Leslie Groves, who Oppenheimer worked for, who hired Oppenheimer.
He asked around, researched, then interviewed Oppenheimer for someone to manage the project’s technical side, then like any good CEO will do, got the hell out of the way and let the work move forward, giving deadlines and making only 20 mile high and inch deep decisions.
Learned a hell of a lot about all levels of a major enterprise when I went into IT security, from the senior most executive level to end user levels (with the technical, from development, through enterprise support, through end user support being within my initial experience base). Whenever someone attempts to operate outside of their scope of experience and knowledge, things turn into a gagglefuck. If a CEO tries to hire a programmer, it’s quite likely that that programmer will have trouble getting HELLO WORLD to work. If a programmer tries to hire a CEO, well, that’ll be another fine debacle. I’ll refrain from considering interviewing CEO types, but I can give reasonable input for programmers and better, know quality programmers that are current in fields close enough to competently interview said programmer candidates. I do have one uncommon skill, recognizing a bullshitter when I see and hear one and some CEO types, typically successful ones in startups have that skill as well.
Not a clue for certain if Altman has that skill, but he’s a salesman at heart, as was evidenced in his testimony that’s in the comments above.
Some time back, in a highly heavily task loaded environment, I had to write filters in regular expressions, with an option to simply go Boolean. I went with RegEx, despite a modest increase on load. One could write the equations in one of two manners, one would massively increase the load, as it had multiple elements to parse and move on from, the other could continue with segments of group conditionals and massively lower loads on the cluster. Given that processing times for existing tasks was in excess of 24 hours, I obviously went for the lower load and hence, faster processing. Some didn’t and I eventually got tasked to review their equations and get the load back down to 24 hours to process the tasks, as it had jumped to 48 hours, rendering all work meaningless. Took about 20 minutes, from first view to final test and code review followed swiftly.
I then took the offender aside and gave a quick and detailed class in how parsing worked, how the system got hopelessly bogged down and how to do the same task without overloading things, for which I was voluminously thanked. Secondary task was to ascertain if said offender would be retained, which he was. A simple misunderstanding of how the logic flowed and loads and we bounced ideas on how to write a minimally intensive test between us. Hell, I learned a new trick, so did the offender, who became one of our better writers.
And quickly enough, I’d ask for peer review on my tests and vice versa.
One reason I studiously avoid higher management, one loses the job of mentoring, along with learning those new tricks.
OK, I did wait, so as not to derail.
Seems to me that his skills according to PZ are evident to anyone who reads this passage.
The skills are those that allow one to acquire lots of money by becoming best buddies with people like Peter Thiel thus making lots of people want to say nice things about him in hope.
(Money is not a skill, of course — but making $$$ surely is)
—
Now and then, I see one of these “tall poppy” posts and wince a tad.
Ah well, no heroes.
John, let’s short circuit it by a lot and realistically.
Altman, Rooby, rooby roo, raggy, a rooby rack!
Zero technical, zero knowledge, plenty of snake oil salesman, nary a clue.
Typically found in used car salesmen.
Just give the fucker a Scooby snack and send him on his way.
wzrd1, heh.
Still, I think PZ understands the concept of an organism thriving within its environment by successfully exploiting a niche. One might not like how it does it, one may think it requires no skill, yet clearly it works.
It occurs to me my allusion might have been a tad too localised, so:
https://en.wikipedia.org/wiki/Tall_poppy_syndrome