[PostOnce]

That's all well and good, but it's not well and good to change what you say because the calculus of what will personally benefit you has changed. It's not the same as "new information coming to light", it is instead greed.

We've already seen a charity robbed to no end other than Profit. Commercializing OpenAI was for one reason: to enrich the people commercializing it. Not to benefit society, and with nary a thought given to the consequences.

Control of AI serves one purpose, in Altman's mind, I am convinced: to prevent other people from eating the chickens he counted too early.

If ten million people lose their jobs to ChatGPT, it's fine as long as Altman & Co grow richer, but if those same people try to democratize AI and make his time and investment valueless in the face of a commoditized category of software, all the sudden he has a problem and those people should be constrained by the government -- but Altman should not?

This won't be the first time greed failed.

[jstummbillig]

I am astonished how anyone could just make a statement like this, here. It's so sticky, it's so damning and brash. It offers absolutely no proof, while contradicting what sama says themselves.

And we are entirely fine with that. Top of the comments after 7 hours.

Note, I am not claiming it could not be true. That is entirely beside my point. The issue is, that it could be entirely untrue. And I really mean, every single piece of information in this comment could be off. And we are comfortable with that. We are burning witches again.

If this is the level of our engagement, in this relatively good and on-topic forum, in times of change and disagreement, I do truly wonder if there is any hope for us.

[PostOnce]

I can think of exactly zero other reasons to have engineered the privatization of a charity, other than greed.

The balance of probabilities indicates to me that the attempt at regulatory capture is also based in greed, but if you have a different view, I'd love to hear your logic.

After all, if AI proliferation were as dangerous as Altman seems to claim, then why is the answer "OpenAI should continue to provide unfettered global public access to quasi-strong AI, but the government should slow other people down "?

If billions of dollars are involved, your first suspicion shouldn't be altruism.

Edit: OK perhaps a valid reason for a private subsidiary is to provide the charity with the income it needs to operate if it has no endowment (OpenAI does), but given all the other points I raised, I don't think that's what's going on here.

[jstummbillig]

> I can think of exactly zero other reasons to have engineered the privatization of a charity, other than greed.

The question has recently been asked by Lex Fridman and answered by Sam Altman. This is the clip: https://youtu.be/qQdqFZFZQ6o

To quote: "We learned early on that were gonna need far more capital than we were able to raise as a non-profit" (you can listen in on it being expanded over 3 Minutes)

Now, since I can almost feel the goal-post-shifting coming, this seems like a good time to align: You claimed you could see no reason. Their reason is: They had a model. Then they figured it was wrong. They adjusted their model.

This is not an insane or unreasonable process. This is how we know things get adjusted all the time. We agree with it on principle.

Now, of course, there are a lot of possible, unfavorable interpretations, even if we can agree up to here. A couple:

a) Despite the process making sense, that's not what happened. Sam might be lying or deceiving. He may be leaving things unsaid, or it could all be a long-planned con, etc. All of this might be entirely possible, and I did not do the work to disprove this option. BUT: as a general rule, we always agree that the burden of proof is on the accuser, in court and in civil discourse. Walking up to people and saying "You stole!", them being irritated, and you saying "Well! Prove me otherwise!" is just not how we do things. I am aware that in practice we often are willing to apply different rules to the very poor, the very rich, or people we just dislike.

b) Sam is incompetent or delusional. Someone else could have made it work without going the route they did. Maybe it would have worked, and he is just... bad at math? Certainly a possibility. Similar to the above, if you want to make this claim without discrediting yourself: Show your work. Reason us through it.

c) This move was illegal. I am not a lawyer, but since I wouldn't rate this thought as particularly groundbreaking, I am okay with assuming someone raised the concern during the transition, lawyers were consulted, and it was a legal option.

d) It's simply immoral to go on, at best the exploitation of a loophole, at worst the biggest treason to humankind. If "open" cannot mean "open source" anymore, before making this transition, letting the company fail was the right move. This is an entirely understandable position, and I can empathize with the feelings it triggers. It requires no evidence, and it requires no claims about greed. It's an opinion piece. Fair enough.

[PostOnce]

I watched the video.

I'll preface this by saying I am neither judge nor prosecutor, and we are not at trial.

Where we are is in a position of considering, before it is too late, the honesty and motives of a man who has taken a charity private because they "needed money".

Needed money for what, exactly? Well, the mission of OpenAI, according to https://openai.com/about their mission is Our mission is to ensure that artificial general intelligence benefits all of humanity.

So how does this newfound capital help that mission? By hiding the details of GPT4, so that no one outside of OpenAI can benefit. I don't even know how many parameters the damn thing has or what kind of growth trajectory we are on, and neither does anyone outside of OpenAI.

Here's a man who stands to control a large part of what resembles AGI, financially benefit to an almost indescribable degree, and control, arguably, a large part of the direction of society (given the number of jobs this will cost), especially if the regulators limit access by Mere Mortals. They won't even tell us the basics of their new models (and newfound powers). So much for "Open" and "benefiting all of humanity".

If someone stands to gain this kind of money and power, and does things that on their face seem dishonest and inimical to society at large, and to the goals of the very charity he privatized, should he be immune to questioning and suspicion?

I am suspicious. I think his actions merit suspicion. I think you all should be suspicious too, based on the events so far.

[jstummbillig]

> So how does this newfound capital help that mission?

Again, as soon as we jump to the "so...", we are already past the point that I was taking offense with. I myself listed a few options from where to take this and you added a few more. Fair enough and entirely fine with me.

This is the point:

> Commercializing OpenAI was for one reason: to enrich the people commercializing it. Not to benefit society, and with nary a thought given to the consequences.

Making bad faith statements about someone without even rebutting the stated intent of the accuse, or even putting in the effort to learn about the intent, is just so incredibly lame and bad style.

Imagine saying this to someones face: "You made this business for one reason alone: To enrich yourself personally, you absolutely do not care about benefitting society, and you don't give a fuck about the damage it might do." Would you think it really be okay to advance this to the accusse, before even bothering to find or hear an explanation regarding the issue you are taking such strong offense with? Just an inkling of what their explanation is without you assuming it for them, as a matter of human decency?

If we cut out the option for people to explain themselves or reason on basis of their explanation, what are we doing other than advancing populism? I really want to believe this place can not be that.

(I want to add, that I have 0 allegiance or interesting opinions about Sama. In the past I found some of their takes interesting, but for me that's just normal, when I listen to a decently smart person for the first time, even if I end up disagreeing with most of their values.)

[PostOnce]

My skepticism is not bad faith, it's honest inquiry based on the facts available to me, which I have enumerated for you.

The hiding of scientific data from a charity has VERY few kind explanations and a lot of terrible ones. The balance of probabilities points towards Sam Altman acting maliciously and not in the public interest, which might be "okay" if he hadn't robbed a charity to do so.

I am still open to contrary viewpoints, but you're not presenting any, you are only saying I am acting in bad faith for what I (and others here) consider to be merited and valid skepticism based on recent events and statements from Altman himself.

Can you provide any charitable interpretation of the hiding of scientific data from a charity whose mission to ensure that artificial general intelligence benefits all of humanity?

[hitekker]

You're in the clear. Like another commenter in the same thread mentioned, you've responded in good faith to a semi-paranoid question. In my uncharitable view, the GP seems to both have a bias for Sam and a weird anxiety over moral questions.

[ShamelessC]

Chill out a little bit. The commenter made it clear that it's their opinion. In fact, they seem to have gone out of their way to suggest it's their opinion - despite the rules here requiring the assumption of good faith (a rule you're breaking, in my opinion).

That it happens to also be a popular opinion doesn't indicate that people are burning witches - it indicates that people are increasingly finding it easier to explain the world in terms of greed and cynicism; because, frankly, that is the world we live in (aghem, in my opinion).

That you find this to be "witch-burning" indicates to me that you must really relate with the ultra-wealthy which honestly just sounds tone deaf. Sam Altman nor Jeff Bezos will "go down" because of a comment on the internet. Do you know who _doesn't_ have that guarantee? Normal people.

[killdozer]

Seriously, we're talking about a company that Microsoft invested 10 billion dollars into, this is very powerful technology that will affect our lives. They didn't invest all that money because it will make the world a better place, they did so because they see the opportunity to make a LOT of money and have the ability to steer where this technology goes while also reducing transparency. Why shouldn't we view that with suspicion, you don't have to roll over just because there's a lot of money involved or because some millionaire/billionaire might get treated "unfairly".

[hutzlibu]

"We are burning witches again."

Excuse me, but what witch is being burned for the wrong reason here?

This debate evolves around Sam Altman, how what he and OpenAI was doing was open - and now it is not anymore, despite contrary initial statements. That is a legit reason for anger, no?

[cardosof]

Burning witches? Call me a radical but I can't help but notice the name of the company is OpenAI, but it's not open at all!

[mnky9800n]

Tbh the only thing I find irritating about the article is the dumb story about being 8 and realizing his future already of being the leader of an ai company.

[femiagbabiaka]

The discourse about lost jobs is necessary but a little odd: leveraging VC money and software driven automation to take over sectors ripe for the picking is the startup model in a box.

[smirth]

I think the point is that society and our economies and particularly people in places of power are not preparing for a world in which automation, AI, and robots rapidly advance to the point where jobs will no longer be needed in many cases, and perhaps one day be a choice not a necessity to even live. It would be better if we prepared for that instead of worshipping jobs as if the point of life is to have a job. The USA is particularly unprepared for this, despite having plenty of wealth to do better. Imagine a workforce half of what we have today. Then figure out a way to have an even better standard of living for everyone. That’s the mission, and it could be possible, unless we measure success of leadership solely by jobs created. People freed from the burden of work would be such a better measurement.

[notjulianjaynes]

To anyone interested in this topic I highly recommend listening to this lecture John Cage gave at Stanford University in 1992.

John Cage - Overpopulation and Art

https://www.youtube.com/watch?v=WzPneYqBLAI

[l8rlump]

Machines can do the work, so that people have time to think.

https://youtu.be/G3V2n9QtpfE

[selestify]

What indications are there that the US will start preparing for this anytime soon? We’ve had decades to share the cost savings from outsourcing to China and failed to do so.

On the other hand, unemploying a large enough segment of society at once might finally force some change in the system, rather than letting it continue limping on in a slow burn where 1-2% of jobs get automated a year.

[bryanrasmussen]

>On the other hand, unemploying a large enough segment of society at once might finally force some change in the system, rather than letting it continue limping on in a slow burn

Proof of employment required to vote.

[edgyquant]

This shows how little you understand the actual source of power in the world. The vote is a way to keep that source from changing governments, not the actual source.

This mindset is prevalent today, as if people are incapable of changing government without its permission

[bryanrasmussen]

Thanks brilliant person on HN for correcting me! I see now that repressive measures are never passed in order to deal with what people in power consider destabilizing or problematic issues.

[noduerme]

And breed.

[pffft8888]

Altman is v2.0 of Zuckerberg, with 1000X more parameters.

I'm worried about v3.0

[EVa5I7bHFq9mnYK]

As an enterprise, they have absolutely no moat. Google, FB or anyone else can come up with a larger system in two-weeks time, and they can well open source it.

[bsaul]

that was my intuition as well, but it seems pretty obvious by now that the tech is a little harder to reproduce thzn that.

otherwise wouldn't have they already done it by now ?