OpenAI to become for-profit company
reuters.com
reuters.com
My understanding: OpenAI follows the same model Mozilla does. The nonprofit has owned a for-profit corporation called OpenAI Global, LLC that pays taxes on any revenue that isn’t directly in service of their mission (in a very narrow sense based on judicial precedent) since 2019 [1]. In Mozilla’s case that’s the revenue they make from making Google the default search engine and in OpenAI’s case that’s all their ChatGPT and API revenue. The vast majority (all?) engineers work for the for-profit and always have. The vast majority (all?) revenue goes through the for-profit which pays taxes on that revenue minus the usual business deductions. The only money that goes to the nonprofit tax-free are donations. Everything else is taxed at least once at the for-profit corporation. Almost every nonprofit that raises revenue outside of donations has to be structured more or less this way to pay taxes. They don’t get to just take any taxable revenue stream and declare it tax free.
All OpenAI is doing here is decoupling ownership of the for-profit entity from the nonprofit. They’re allowing the for profit to create more shares and distribute them to entities other than the non-profit. Or am I completely misinformed?
[1] https://en.wikipedia.org/wiki/OpenAI#2019:_Transition_from_n...
Obviously Elon is mostly doing this suit as a way to benefit Grok AI but honestly I don’t mind that; competitors are supposed to keep each other in check, and this is a good and proper way for companies to provide checks & balances to each others’ power and it’s one reason why monopolies are bad is the absence of competitor-enforced accountability.
Lawsuit: https://www.reuters.com/technology/elon-musk-revives-lawsuit...
https://www.reuters.com/technology/elon-musk-revives-lawsuit-against-sam-altman-openai-nyt-reports-2024-08-05/> how this isn't the case here.
Its not the case because they are doing the opposite of what you are suggesting. They are increasing the value of the asset that they own.
Sure, the asset itself is being diluted, but the individual parts that it owns are more valuable.
It is perfectly reasonable for a non profit to prefer to own 30% of a 100 billion dollar asset, lets say, compared to 100% of a 10 billion dollar asset.
The goal of the openAI non-profit is something something control the development of AI for the good of all humanity, then it seems that they explicitly shouldn't care about making $20 billion, and explicitly should care about maintaining control of openAI.
If you listen to their rhetoric, $20 billion is peanuts compared to the lightcone and the kardashev scale and whatever else.
Yes, and if you have a bunch more money then you can do more non profit activities that help the world.
Getting as much money as possible, so that the money can be used for your great cause, is the best way to effectively run a non profit.
> then it seems that they explicitly shouldn't care about making $20 billion
Of course they should, because that 20 billion dollars can be used for its goal more effectively than having control over a lower value asset.
> compared to the lightcone and the kardashev scale and whatever else.
You are pre-supposing that openAI's model itself is some magic, infinitely valuable asset already.
Its not. If it were, then it would already be worth 10 trillion dollars. But its not worth that.
Therefore the money is worth more than the asset. There are lots of other AI groups around here. OpenAI is just one of them, and they are not infinitely valuable.
> Getting as much money as possible, so that the money can be used for your great cause, is the best way to effectively run a non profit.
In anywhere but Silicon Valley is a great way to violate Unrelated Business Income limits and get your charitable status revoked. It is not sufficient that a non-profits "goals" be charitable, their day-to-day activities must be as well, and it's not acceptable to put off those activities until some future date when you'll "make up" for all the regular for-profit work.
Good thing this wouldn't be that. Instead, it would be about promoting the cause.
And yes, non profits are allowed to own assets and maximize the value of those assets.
Of course their mission also matters and they should push towards that. But throwing away billions and billions of dollars for nothing isn't the way to do that.
> their day-to-day activities must be as well
Yes.... and they should also do that.
That has absolutely nothing to do with refusing to sabotage your non profit by throwing away a bunch of money for no reason though.
Of course the non profit should work towards their goal in their day to day activities.
> until some future date
Who said anything about waiting for a future date? Of course their current actions should push towards their goal.
That still has nothing to do with refusing to set money on fire for no reason though.
If anything, I think that the people who were attempting to set their valuable assets on fire and sabotage the non-profit are the ones who should be prosecuted by the legal system to the fullest extend legally allowed for going against the mission and intentionally engaging in charity fraud.
At one point, some of those board members said something about how that they were seriously considering shutting the whole thing down. I would absolutely consider that to be extremely illegal charity fraud, deserving of jail time if they did that.
I don't think that's true? A non-profit can sell products or services, it just can't pay out dividends.
What counts as “related” to the charity’s mission is fuzzy but in practice the courts have been rather strict. They don’t have to form for-profit subsidiaries to pay those taxes but it helps to derisk the parent because potential penalties include loss of nonprofit status.
For example, the nonprofit Metropolitan Museum of Modern Art has a for-profit subsidiary that operates the gift shop. National Geographic Society has National Geographic Partners which actually owns the TV channel and publishes the magazine. Harvard and Stanford have the Harvard Management Company and Stanford Management Company to manage their endowments respectively. The Smithsonian Institute has Smithsonian Enterprises. Mayo Clinic => Mayo Clinic Ventures. Even the state owned University of California regents have a bunch of for-profit subsidiaries.
As you've realized, this should have been (and was) obvious for a long time. But that doesn't make it any less hypocritical or headline worthy.
In terms of profit, AFAICT, Sam doesn't have designs on building extra large yachts and his own space agency but what he wants is to be the one at the stead of building what he considers is world-changing tech. One could rationally call this power-hungry but one could also rationally call this just helicopter parenting of a tech you've helped built. And for that a for-profit that is allowed to maximize profits to re-invest in the tech is the optimal setup (esp if all the competitors are doing the same)
Is this a different org than when it started? Yes. Was this a dupe from the beginning? I don't think so.
"But why can't he have a more worldly-aligned board looking over his shoulder?"
Because we live in California and have a bad taste for governance by committee or worse: governance by constant non-representative democracy (see: Housing).
If this now completely comes off the wheels, I still think Congressional action can be a stopgap, but atleast for now, this restructure makes sense to me.
When seen through the rearview mirror, the whole narrative screams of self-importance and duplicity. GPT-2 was too dangerous, and only they were trust-worthy enough to possess. They were trust-worthy because this was a non-profit, so "interest aligned with humanity". This charade has continued even to barely some months ago.
World: "But what if it was like, a lot of money?"
Sam: "Oh alright you convinced me. Fuck my principles."
I could tell you in very plain terms how a competitor to Boeing and SpaceX would benefit the American economy. I have not even the faintest fucking clue what "AGI" even is, or how it's profitable if it resembles the LLMs that OpenAI is selling today.
People use SpaceX every day even if they never connected to a starlink -- the lower costs that governments pay for space launches means more money for other things, not to mention no longer paying Russia for launches or engines.
Which sounds all well and good until you realise it’s at the complete whims of one highly misinformed and reactionary individual.
He’s one made-up article away from turning sides and fucking everything up.
https://en.wikipedia.org/wiki/Starlink_in_the_Russo-Ukrainia...
https://en.wikipedia.org/wiki/Views_of_Elon_Musk#Russian_inv...
On the other hand, just because the execs who do the layoffs bought into the narrative it doesn't mean they're right.
what annoys you?: https://www.youtube.com/watch?v=k9Le1ibX2zY
> if you went to a group of investors and pitched a board game where the winners get space ships and the losers die, they'd call you crazy. But if you suggested to those same investors that perhaps we shouldn't organize our entire society that way, they'd call you crazy.
The Social Security budget is $1.4 Trillion, just the federal welfare budget is >$1Trillion (not including state budgets), and then there's medicare. Meanwhile, the NASA budget is <$25B (with SpaceX's operating budget and profits being a fraction of that)
I wish we lived in that simple of a world. But we don't.
Was the original quote. I just showed that you could get rid of all rockets ever made and it would be a rounding error for any funds used to save the “losers” from dying.
How is that a non-sequitur?
Feed the hungry. House the homeless. Give away money unconditionally to those in need. Build hospitals in poor countries. Fight disinformation on crucial topics (such as climate change). Provide disaster relief. Not build more power hungry technology that exacerbates our current problems.
Do literally anything positive for another person, that does not harm others.
The list is pretty big when one isn’t selfish; there’s no law forcing anyone to build space agencies.
A lack of imagination is not an excuse.
Funnel $10B in housing to Los Angeles and you'll build less than 100 units of housing, because the inflationary push of that money would balloon the cost of per unit housing. I don't want to imagine the effect of that on middle class housing.
Funnel $10B of food to xyz famine region and you've undercut local farmers for generations. Happens all the time [1]. And that's assuming you can get the aid past local corruption.
These problems aren't as simple as people assume, and I'm low-key happy young naive Billionaire's are avoiding these issues instead of trying to throw their weight around.
FWIW: Sam's already funneled a bunch of money into green energy production[2].
[1]: https://haitisolidarity.net/in-the-news/how-the-united-state...
[2]: https://www.cnbc.com/2024/09/25/sam-altman-backed-nuclear-st...
Doesn't make sense to me. An uptick in construction work will not be an inflation balloon. More disposable income doesn't mean 1:1 more spending.
If you build a lot of (social) housing, you put at worst a lot of people a roof above their head.
Families having less financial stress might lower crime rate and improve children school scores. They might save to start businesses or find their other talents.
For some, this might be a downside tough. It makes workers more educated, healthier, more stable, less desperate and less dependent on bosses, plus they might be less angry so politically less exploitable too.
There's a massive shortage in construction workers [1], so yes there will be? The few construction workers we do have can demand higher wages (yay!) but then will they be outbidding other mid-income folks for housing with those increased wages? Sounds like an inflation spiral to me.
My statement wasn't against social housing, I love social housing. We just haven't cracked the code in scaling housing (and subsequent maintenance) yet. And the problem is about 80% political will, billionaire cash is useless here.
[1]: https://www.abc.org/News-Media/News-Releases/abc-2024-constr...
It is rather the other way around. Higher rents / house prices will make sure only people with higher wages can afford to live there. That means your bagel or coffee will be more expensive there too.
Pretty much everything required to build housing, wood, labor, pre-approved land is in a massive shortage that we can’t spend money to fix.
So more money to simple pump demand for all those things will have a massive inflationary impact.
Please, forget anything you are worrying about here, it does not apply.
If just these problems could be solved the state has more than enough funds to house everyone. What billionaires do would be wholly irrelevant (like it is now)
[1] https://www.constructiondive.com/news/construction-materials...
Start by collaborating with organisations which are entrenched in studying these issues and the impact of the solutions. If you have the money you can pay them to help and guide the effort, don’t act like if you know everything.
They won’t touch American housing problems with a 10ft pole. That should tell you something.
Go to Berkeley, tell them a Billionaire wants to build housing for the homeless in their neighborhood. See what happens.
It’s a hard pill to swallow but the best thing billionaires can do is let us tax them and then butt out go fly rockets. The political problems is upto the rest of us.
Why do you keep insisting on the USA? It’s not the only country in the world.
> Go to Berkeley
I will not. I’m not American.
> It’s a hard pill to swallow but the best thing billionaires can do is let us tax them
Maybe it’s a hard pill to swallow for the billionaire, but I personally agree and think you’re right. However, this conversation started with someone asking “what do you do with a a lot of money past a point” and offering only a private space agency as an alternative to working on AGI. My point was there are many other problems worth pursuing.
If you don’t think NIMBYism and degrowth is a problem in your country yet, just give it a couple of years. It just hit England, you’re next. No billionaire can save you.
That's not least complexity. That's least effort.
And now they want to cast off any pretense of that former altruistic yolk now that they have a new, better raison d'etre to attract talent: making absolutely unparalleled stacks of cash.
The board of the non-profit fired Altman and then Altman (& MS) rebelled, retook control, & gutted the non-profit board. Then, they stacked the new non-profit board with Altman/MS loyalists and now they're discharging the non-profit.
It's entirely about control. The board has a legally enforceable duty to its charter. That charter is the problem Altman is solving.
We worry about the existential threat of AI becoming a paperclip factory while many humans are already.
Right now, OpenAI, Inc. (California non-profit, lets say the charity) is the sole controlling shareholder of OpenAI Global LLC (Delaware for-profit, lets say the company). So, just to start off with the big picture: the whole enterprise was ultimately under the sole control of the non-profit board, who in turn was obligated to operate in furtherance of "charitable public benefit". This is what the linked article means by "significant governance changes happening behind the scenes," which should hopefully convince you that I'm not making this part up.
To get really specific, this change would mean that they'd no longer be obligated to comply with these CA laws:
https://leginfo.legislature.ca.gov/faces/codes_displayText.x...
https://oag.ca.gov/system/files/media/registration-reporting...
And, a little less importantly, comply with the guidelines for "Public Charities" covered by federal code 501(c)(3) (https://www.law.cornell.edu/uscode/text/26/501) covered by this set of articles: https://www.irs.gov/charities-non-profits/charitable-organiz... . The important bits are:
The term charitable is used in its generally accepted legal sense and includes relief of the poor, the distressed, or the underprivileged; advancement of religion; advancement of education or science; erecting or maintaining public buildings, monuments, or works; lessening the burdens of government; lessening neighborhood tensions; eliminating prejudice and discrimination; defending human and civil rights secured by law; and combating community deterioration and juvenile delinquency.
... The organization must not be organized or operated for the benefit of private interests, and no part of a section 501(c)(3) organization's net earnings may inure to the benefit of any private shareholder or individual.
Naturally, this makes it very hard to explain how the nonprofit could give up basically all of its control without breaking its obligations.
All the above covers "why does it feel unfair for a non-profit entity to gift its assets to a for-profit", but I'll briefly cover the more specific issue of "why does it feel unfair for OpenAI in particular to abandon their founding mission". The answer is simple: they explicitly warned us that for-profit pursuit of AGI is dangerous, potentially leading to catastrophic tragedies involving unrelated members of the global public. We're talking "mass casualty event"-level stuff here, and it's really troubling to see the exact same organization change their mind now that they're in a dominant position. Here's the relevant quotes from their founding documents:
OpenAI is a non-profit artificial intelligence research company. Our goal is to advance digital intelligence in the way that is most likely to benefit humanity as a whole, unconstrained by a need to generate financial return. Since our research is free from financial obligations, we can better focus on a positive human impact...
It’s hard to fathom how much human-level AI could benefit society, and it’s equally hard to imagine how much it could damage society if built or used incorrectly. Because of AI’s surprising history, it’s hard to predict when human-level AI might come within reach. When it does, it’ll be important to have a leading research institution which can prioritize a good outcome for all over its own self-interest.
We commit to use any influence we obtain over AGI’s deployment to ensure it is used for the benefit of all, and to avoid enabling uses of AI or AGI that harm humanity or unduly concentrate power. Our primary fiduciary duty is to humanity...
We are concerned about late-stage AGI development becoming a competitive race without time for adequate safety precautions. Therefore, if a value-aligned, safety-conscious project comes close to building AGI before we do, we commit to stop competing with and start assisting this project. We will work out specifics in case-by-case agreements, but a typical triggering condition might be “a better-than-even chance of success in the next two years.”
Sorry for the long reply, and I appreciate the polite + well-researched question! As you can probably guess, this move makes me a little offended and very anxious. For more, look at the posts from the leaders who quit in protest yesterday, namely their CTO.
Look at your OpenAI invoices. They're paid to OpenAI LLC, not OpenAI Inc. I can't find confirmation on openai.com what the exact relationship between OpenAI Global LLC and OpenAI LLC is but the former is on their "Our Structure" page and the latter is in their data processing addendum so it's probably the subsidiary in charge of operating the services while Global does training and licenses it downstream. OpenAI Global was the one that made that big $10 billion deal with Microsoft
That obligation is why they had to spin off a for-profit corporation. Courts are very strict in their interpretation of what "unrelated business income" is and the for-profit LLC protects the non-profit's tax exempt status.
> "why does it feel unfair for a non-profit entity to gift its assets to a for-profit"
What assets were gifted, exactly? They created the for-profit shortly after GPT2 (in 2019) and as far as I can tell that's the organization that has developed the IP that's actually making money now.
I honestly don't understand how this isn't in the interest of the nonprofit's mission. It's currently a useless appendage and will never have any real power or resources until either OpenAI is in the black and sending profit up to it, or they can sell OpenAI shares. I don't think the charity has any more claim over GPT4 than Google does, having invented transformers.
If this next round of funding goes through at $100-150 billion valuation, OpenAI Inc will probably be (on paper at least) the second wealthiest charity on the planet after the Novo Nordisk Foundation. This restructuring opens the way for the nonprofit to sell its shares and it's going to be a hell of a lot of money to dedicate towards their mission - instead of watching its subsidiary burn billions of dollars with no end in sight.
What assets were gifted, exactly? ...It's currently a useless appendage and will never have any real power or resources until either OpenAI is in the black and sending profit up to it, or they can sell OpenAI shares.
This restructuring opens the way for the nonprofit to sell its shares and it's going to be a hell of a lot of money to dedicate towards their mission
Finally, one tiny side point:
Courts are very strict in their interpretation of what "unrelated business income" is and the for-profit LLC protects the non-profit's tax exempt status.
Again, I'm a noob, so I'll cite the IRS-for-dummies page on the topic for onlookers: https://www.irs.gov/charities-non-profits/unrelated-business...
That's not my interpretation. It sounds like the restructuring is part of the deal that values OpenAI at $100-150 billion [1]. They're not "gifting" away anything any more than a Series A investor gifts something to a Series B investor. They're restructuring so that their stake will be worth more afterwards than it is now, regardless of the percentages. That's what every company owner goes through when the company raise a VC round, goes public, or even just offers employees stock options. That doesn't change because the owner is a nonprofit and it sounds like until they restructure, the for-profit is worth nowhere near that crazy 12 figure number.
"Minority stake" just means that they won't have enough to control the corp out right with 50%+1 which is probably what everyone wants to justify the investment. Reuters TFA says "The plan is still being hashed out with lawyers and shareholders and the timeline for completing the restructuring remains uncertain" so we don't really know what the post valuation numbers look like or who is getting what. We also don't know how the voting vs non-voting shares will split. Losing majority control after multiple multi-billion dollar rounds is the norm so if Microsoft's previous $10 bil investment converts and 10-20 pts go to the employee pool, it's a perfectly fair deal.
> Even putting aside the core issue above (they won't have many shares to sell), the second part of my comment comes back here: what would they buy with all that money? Anthropic? Their explicit mission is to beat for-profit firms in the race to AGI so convincingly that an arms race is avoided. How could they possibly accomplish this after selling control of the most capable AI system on the planet?
As far as I can tell, without this deal OpenAI LLC goes bankrupt under the rumored $5b/yr losses and the charity loses all relevance when the bankruptcy court fire sales the IP. With this deal, it can create a secondary market and use the funds to focus on its actual mission. The only way that his deal doesn't make sense (in my mind) is if you believe that GPT4/o1/whatever are the keys to fulfilling OpenAI's mission and it becomes impossible if it loses control. Personally I find that very hard to believe, which might be the actual disconnect we're having here.
What would they buy? They'd hire people to do research under their umbrella instead of a for-profit one, fund compute infrastructure for researchers who don't have billions for H100s, give grants to organizations, or spin off more startups. Even if it's a 20% stake of a $100 billion, that's enough for a ivy league sized endowment that can fund AI research for generations. If it's a 49% stake of $150 billion, that makes it the second wealthiest charity after Novo Nordisk Foundation - which continues to do tons of biomedical research even though it doesn't have total control over the public Novo Nordisk or the semaglutide IP.
OpenAI would become one of the largest grant giving organizations in the world overnight using just the interest from the endowment. Imagine the equivalent to 50-100% of the NSF's annual grant budget going just to AI research!
> Normal charities have no issues with unrelated business income, because they don't run "trades or businesses", they just spend their money.
See my other reply [2] for a sample of some nonprofits that have for-profit arms. A significant fraction of them do, especially those that offer some sort of product or service (I don't have hard stats but I'd venture it's the majority of the major charities that do the latter). "Normal" charities that just disburse grant money and spends donations is only one type among many.
[1] https://www.reuters.com/technology/artificial-intelligence/o...
I’d say we disagree about the following somewhat indeterminate points:
1. Whether OpenAI has/could-have-had staying power without raising immense amounts of venture capital. I will readily admit that they’ve gone so far down this road that they are now somewhat trapped by their massive investments and contracts, not to mention losing almost all their top researchers.
2. Whether the people in charge of this deal can be trusted to propose a fair outcome for the charity other than “their mission ineffably lives on in us” (and as a corollary, whether a fair outcome is likely).
3. Whether the private technical assets of OpenAI (GPT, DALLE, and Sora) are meaningfully unique in their potential for impact, knowing what the public knows in the current moment — which I will admit is far from the complete competitive picture.
4. Whether OpenAI’s mission could be meaningfully achieved by passing out grants to a diverse body of scientists.
I’d be happy to “debate” (lol) any of those particulars if you want, but I think it’s otherwise best to leave it at “we assess the known facts differently”. If we let some time pass, that’ll at least settle question 2…
Since Sam Altman is on the board of OpenAI Inc, I expect this deal will be under extreme scrutiny for self dealing. He has flown under the radar so far by not taking any equity but this changes the second he does (IANAL). California, Delaware, and the feds will be looking closely at the deal.
I don't think the danger for OpenAI the charity is as great as people make it out to be. They'll be able to do a lot more than hand out grants with an 11 figure endowment.
Yes, but going from being controlled by a nonprofit to being controlled by a typical board of shareholders seems like a pretty big change to me.
In this case, Mozilla as a non-profit owning a for-profit manages to more or less fulfill the non-profit's mission (maintaining an open, alternative browser). OpenAI has been in a hurry to abandon it's non-profit mission for a while and the complex details of its structure doesn't change this.
All? As far as I know this is unprecedented.
But unfortunately charities and not-for-profits putting their core business into a company, and then eventually selling it off is not unprecedented. For example, The Raspberry Pi Foundation was a not-for-profit organisation around Raspberry Pi. They formed a LLC for their commercial operations, then gradually sold it off before eventually announcing an IPO: https://www.raspberrypi.org/blog/what-would-an-ipo-mean-for-....
I think it is terrible that not-for-profits are just being used as incubators for companies that eventually take the core mission and stop primarily serving the public interest.
There are of course other examples of charities or not-for-profits that put part of their core operations in a company and don't sell out, instead retaining 100% ownership - for example Mozilla. However, I think there should be some better way for impactful not-for-profits to have some revenue generating aspects in line with their mission (offset by allowing temporary surplus to cover future expenses, or by other expenses).
I don't think it's unprecedented, even at this scale.
Novo Nordisk, the pharmaceutical company behind Semaglutide (aka Ozempic) with a market cap >$600 bilion, was founded by the Novo Nordisk Foundation before going public. The latter now has an endowment of over $150 billion and owns a significant fraction of the public company.
The nonprofit invests the tax free donations into the for-profit. It gets to keep its equity just like any other investor and as long as that equity held by the non-profit, there's no taxable event. If the nonprofit sells its share - since it was closely involved in the creation and management of the for-profit as an active investment - it becomes a taxable event under the unrelated business income tax rules. Until that time, the only "profit making applications" like ChatGPT and the API are run by the for-profit, which - I repeat - pays its taxes.
I genuinely don't understand why people think they've skirted taxation except out of sheer ignorance of how non-profits actually work. A 501(c)(3) is not some magic Monopoly "get-out-of-tax-free" card and the IRS isn't stupid, there's a ton of rules for tax exemption. They'd have a much easier time with tax avoidance if they were an actual for profit corporation with billions of dollars because GAAP rules are a lot more forgiving than non-profit regulations.
[1] "The Mozilla Foundation has no members" https://hacktivis.me/articles/mozilla-foundation-has-no-memb...
The non-profit could maybe sell its assets to investors, but then what would it do with the money?
I'm sure OpenAI has an explanation, but I really want to hear more details. In the most simple analysis of "non-profit becomes for-profit", there's really no way to square it other than non-profit assets (generated through donations) just being handed to somebody for private ownership.
If that's how it works, why wouldn't you start every startup as a non-profit?
Investment is tax deductible, no tax on profits...
Then turn it into a for-profit if/when it becomes successful!
The biggest problem with this is that there's basically no chance that the sale price of the non-profit assets is going to be $150 billion, which means that whatever the gap is between the valuation of the assets and the valuation of the company is pure profit derived from the gutting of the non-profit.
If this is allowed, every startup founded from now on should rationally do the same thing. No taxes while growing, then convert to for profit right before you exit.
What do you propose should be taxed, exactly?
How do you know they had no profit with all of the deals with major companies and having one of the most popular software services in existence? Non-profits can earn profit, they just don't have to pay taxes on those profits and they can't distribute those profits to stakeholders -- it goes back to the business.
They are also a private company, and do not have to report revenue, expenses, or profits.
So yeah, I stand by what I said -- it sounds like fraud. And it deserves an audit.
By reading their Form 990 filings, which are publicly accessible here: https://projects.propublica.org/nonprofits/organizations/810....
Perhaps it would need to be something like x%revenue above $10M, y%revenue above 1B beginning after three years of operation.
> Cost of goods sold (COGS) refers to the direct costs of producing the goods sold by a company. This amount includes the cost of the materials and labor directly used to create the good. It excludes indirect expenses, such as distribution costs and sales force costs.
It pushed almost all SWE jobs to be classified as R&D jobs, which changed how taxes are calculated on companies.
They have an example at [0], but I'll copy it here. For a $1mm income, $1mm cost of SW dev, with $0 profit previously you paid $0 in tax (your income was offset by your R&D costs). Now it would be about $200k in taxes for 5 years, as you can't claim all of the $1mm that year anymore.
In this case, business have to pay taxes on "profit" that they don't have as it immediately went to salaries. There were a lot of small business that were hit extremely hard.
They tried to fix it in the recent tax bill but it was killed in the Senate last I checked. You can see more here: https://www.finance.senate.gov/chairmans-news/fact-sheet-on-....
Also, software developers in Oil and Gas industries are exempt from this :)
The Trump tax policy was a bizarre move for a country that relies so heavily on homegrown innovation. But then again, so was the entire Trump presidency.
The vast majority of taxes paid in developed nations are employee taxes and whatever national+local sales taxes and health/pension equivalent taxes are (indirectly) levied (usually 60-80% of national income). Asset taxes are a bit different.
It's true even in the bootstrapped company case: If you earn say $100k and keep $50k after all the employee indirect/direct taxes. Now imagine you spend $40k of that $50k in savings, setting up a business. You spend $30k on another employee, paying $15k of employer and employee taxes, and spend the other $10k on a company to do marketing (who will spend $5k of that on employees and pay $2.5k of tax), and you earn less than $40k in income, by the end of year 1 you have:
1) A loss-making startup which nonetheless is further along then nothing
2) Out of $100k of your original value, $67.5k has already reached the government within 12 months
3) Your time doing the tech side was not compensated but could not (for obvious anti-fraud reasons) be counted as a loss and as you have noted, you don't pay tax when you make a loss, and you don't get any kind of negative rebate (except certain sales tax regimes or schemes).
If you are in the US, the above is currently much worse due to the insane way R&D Software spend needs to be spread immediately as a tax burden.
So it's really not fair to say a new startup isn't paying taxes. They almost always are. There are very few companies or startups that pay less than 50% of their income to staff, and almost all of those are the unicorns or exceptional monopoly/class leaders. Startups, and founders tend to disproportionately give more of their income and are essentially to that extent re-taxed.
Even though you saved the money in order to start a startup, and paid your due employee taxes, you then have to pay employee taxes to use it, etc.
EDIT: I guess we do have employer tax as national insurance contributions too, always forget about that since I’ve always paid myself under that threshold
The UK does have employers NI contributions but that's not what I mean. The point is, if you spent a year to earn a gross £100k, and as you earn it, pay £50k of total tax, and with the remaining £40k/£50k you spend it on an employee at your company in salary and pay then £20k of tax, the government has that year earned £70k from that £100k passing through.
You can argue that really "£140k" has passed through, but it's not the case, because you created a new job that wouldn't otherwise have existed had you instead saved that £40k for a house. Either way HMRC gets £70k this year rather than £50k.
The wider point I was making is that all companies, even for-profit, pay tax to do just about anything, and companies with much lower sales than costs aren't just paying nothing. They generally have higher costs because they are paying people, and paying their taxes every month. The tax per employee is completely uncorrelated with the financial profit or thereof by the business, so it's a (sensible) misconception that companies that don't make profit like startups don't contribute to the economy. They do, by paying employment taxes.
I'm really making the point that you have to account for employee taxes (both employer and employee as you mention) for your costs as a business. That means, even though you already paid those yourself when you carried out the work to gain savings to invest in your business (to spend on an employee), you have to pay again when paying your employee.
I.e. Self-funded or businesses launched from previous accrued personal income where you invest your own time as well result in a bad tax situation;
whereas an employee earning £100k might pay £50k tax total and save £50k for a house (no VAT),
The alternate of investing that £50k in your business by paying someone £40k means you have to pay that employees PAYE, their Employer and Employee NI. So the government gets to re-tax most of that money when you use it to hire someone to build a new business with you, in a way they don't if you use it to buy a house, in terms of practical impact. When you pay yourself as an entrepreneur depends, there's dividends+PAYE in the UK (which requires yes you pay for both your employer and employee tax for yourself) or capital gains(ignoring tax schemes), either way, you do get taxed at some point to bring cash out.
The government in other words massively benefits from unprofitable for-profit companies so long as they hire some people, especially if the companies are self-funded. But even if it is investment, it's better to have that money spent on salaries now in new companies than sitting as stock in larger companies that keep cash reserves or use schemes to avoid tax. They get much more tax from people starting even unprofitable new businesses, than from employees who simply save money.
It's one of the reasons that since the introduction of income taxes (more or less WW1 in most countries!), you need money to get money in way that you fundamentally did not in the same way back when you could earn $50 from someone and directly use that same $50 to pay someone for the same skills without any loss of value.
You should consider it also from the point of view of the employee. The government taxes your employee to offer him services, it does not care who hires him (you, that saved the money).
Yes, it is true that you need lots of money to HIRE someone, but you can try to do a startup with a couple people that live from their savings for a while (so, not paying themselves a salary, but having shares) which avoids the tax situation as first.
I think we are quite bad to assess how was life around 1900 in terms of infrastructure (in any country) - so yes, probably people paid less taxes but lived in much worse overall conditions.
Forget who the government is supposed to be taxing for what supposed purpose. The decision about asset and income law working differently (liabilities counting for one and all sources being able to intermingle over the financial year, financially speaking, for assets - with income always being payable within a month) is why these taxes work differently in practice then just one being for income, and one for personal asset (accrual). We could instead "tax an employee to offer his services" with a tax which allowed them to discount the liability of savings spent in businesses from their due tax from other sources, or we could charge higher capital gains than personal taxes.
If you earned the original income from renting out properties or capital gains however and then invested it, you can write it off as a loss for your overall individual capital gains, pay $0 for all your rental/share increase in value, and only pay for the startups employee, with no tax on your original income as a result that tax year.
If you have asset wealth, you don't get taxed twice like this as you can write it off. If you have income based savings wealth, you always get taxed and can't count it against investments you make.
1900 is obviously different but income taxes help people with assets retain them for the reasons mentioned above. If Assets were taxed at a higher rate and you could not personally include liabilities in your capital gains (as with income), then it would be the opposite scenario.
We say capital gains tax is all about wealth, but it's not: The US has no wealth tax. The capital gains tax is just lower tax on unearned income and the ability to intermingle that income. It's all income at the end of the day - just one, income from work, the government taxes heavily, the other, the government taxes less heavily, but most people never significantly earn that income.
Of course the investors do end up owning their shares at a lower basis than they would otherwise, and they end up a bit diluted compared to a straightforward investment, but the investors seem likely to more than make up for this by donating appreciated securities to the 501(c)(3) and by deferring or even completely avoiding the capital gains tax on their for-profit shares.
Obviously everyone needs to consult their lawyer about the probability of civil and/or criminal penalties.
Practically the same as selling, but technically not. Non-profit still gets to live up to it's original mission, on paper, but doesn't really do anything internally.
The cleanest way for this to work is the for-profit to just sell more shares at the $150B valuation, diluting the non-profit entity below majority ownership. The for-profit board, which the non-profit could still probably have multiple seats on, would control the real asset, the non-profit would still exist and hold many tens of billions of value. It could further sell its shares in the non-profit and use the proceeds in a way consistent with its mission.
They wouldn't even have to sell that much - I am pretty sure the mega-fundrasing rounds from Microsoft etc brought the non-profit's ownership to just north of 50% anyway.
I don't see how this wouldn't be above board, it's how I assumed it was going to work. It would indeed mean that the entity that controls ChatGPT would now be answerable to shareholders, a majority of which would be profit seeking and a minority of which would be the non-profit with its mission, but non-profits are allowed to invest in for-profits and then sell those shares; all the calls for prosecutions etc seems just like an internet pitchfork mob to me.
The non-profit’s asset is the value of OpenAI minus the value of its profit-participation units, i.e. the value of the option above the profit cap. Thus, it must be less than the value of OpenAI. The non-profit owns an option, not OpenAI.
― L. Ron Hubbard
Every answer moving forward now will contain embedded ads for Sephora, or something completely unrelated to your prompt...
That money will go into the pockets of a small group of people that claim they own shares in the company... Then the company will pull more people in who invest in it, and they'll all get profits based on continually rising monthly membership fees, for an app that stole content from social media posts and historical documents others have written without issuing credit nor compensating them.
As long as the money doesn't go into someone's pocket, it's all good (except that Sam Altman is also getting equity but I assume they found a way to justify that.)
OpenAI will eventually be forced to convert from a public charity to a private foundation and will be forced to give away a certain percentage of their assets every year so this solves that problem also.
It's been many years, but the plan was essentially this:
* The original, non-profit would still exist
* A new, for-profit venture would be created, with the hospital having a board seat and 5% ownership. Can't remember the exact reason behind 5%. I think it was a threshold for certain things becoming a liability for the hospital as they'd be considered "active" owners above 5%. I think this was a healthcare specific issue and unlikely to affect non-profits in other fields.
* The for-profit venture would seek, traditional VC funding. Though, the target investors were primarily in the healthcare space.
* As part of funding, the non-profit would grant exclusive, irrevocable rights of it's IP to that for-profit venture.
* Everyone working for the "startup" would need to sign a new employment contract with the for-profit.
* Viola! You've converted a non-profit into a for-profit business.
I'm fuzzy on a lot of details, but that was the high level architecture of the setup. It's one of those things where the lawyers earn a BOAT LOAD of money to make sure every technicality is accounted for, but everything is just a technicality. The practical outcome is you've converted a non-profit to a for-profit business.
Obviously, this can't happen without the non-profit's approval. From the outside, it seems that Sam has been working internally to align leadership and the board with this outcome.
-----
What will be interesting is how the employees are treated. These types of maneuvers are often an opportunity for companies to drop employees, renegotiate more favorable terms, and reset vesting schedules.
This is the part that should land people literally in jail. A non-profit should not be able to donate its assets to a for-profit, and if it's the same people running both companies, those people must be sent to prison for tax evasion. There is no other way to preserve the integrity of the "non-profit" status with this giant loophole.
Isn't that fraud/stealing from all the donors? I mean how is that different than just giving money to another business not owned by the non profit?
https://www.businessinsider.com/sam-altman-openai-note-more-...
OpenAI has been one of the most insane business stories in years. I can't wait to read a full book about it that isn't written by either Walter Isaacson or Michael Lewis.
"Villain staging the show / open, close / you can count on the con man to wow you / even though, the only trick he knows / is the “law” of scale / but let's just hope / The con man doesn't turn into evil / when the thing he has is real and powerful"
[1] https://www.drmindle.com/ai-is-not-dangerous/#villains-in-th...
List of crawlers for those who now want to block: https://platform.openai.com/docs/bots
https://github.com/ai-robots-txt/ai.robots.txt/blob/main/rob...
cloudflare have a button for this:
https://blog.cloudflare.com/declaring-your-aindependence-blo...
"If I read your book and I have a photographic memory and can recall any paragraph do I need to pay you a licensing fee?"
"If I go through your library and count all the times that 'the' is adjacent to 'end' do I need to get your permission to then tell that number to other people?"
Ironically the one person with resources fighting it in a tangible way, even if for spite, is Elon Musk.
But the board's lack of communication apparently allowed Altman to demonstrate he was more important to the organization than the formal/legal structure, 90% signed intents to quit and the board backed down. It seemed that Altman simply represented the attitude of the many Silicon Valley tech-people - once you have a chance of money, don't hold back, do everything you can to make it.
I can see large copyright holders lining up with takedowns demanding they revise their originating datasets since there will now be a clear-cut commercial use without license.
Hopefully at least in the EU someone will wake up and make better laws or even start applying them to the current situation.
https://www.theparisreview.org/blog/wp-content/uploads/2018/...
I thought so for a moment but then again Meta, Anthropic (I just checked and they have a "for profit and public benefit" status whatever that means), Google or that Musk's thing aren't non-profits, are they ? There are lawsuits in motion for sure but with how it stands today I think ai gets off the hook.
Early hires, who were lured there by the mission?
Donors?
People who were supposed to be served by the non-profit (everyone)?
Some government regulator?
In a normal situation, the primary people with standing to prevent such a move would be the board members of the non-profit, which makes sense. Luckily for Sam, the employees helped kick out all the dissenters a long time ago.
Would the founding documents actually count as proof of a lie? I feel like the defense could easily make the argument that the documents accurately represented their intent at the time, but as time went on they found that it made more sense to change.
It seems like, if the founding documents were to be proof of a lie, you'd have to have corresponding proof that the documents were intentionally written to mislead people.
You’re right that Altman is/will sell it as an unexpected but necessary adaptation to external circumstances, but that’s a hard sell. Potentially not to a court, sadly, but definitely in the public eye. For example:
We are concerned about late-stage AGI development becoming a competitive race without time for adequate safety precautions… We are committed to providing public goods that help society navigate the path to AGI.
And this is the very first paragraph of their founding blog post, from 2015:
OpenAI is a non-profit artificial intelligence research company. Our goal is to advance digital intelligence in the way that is most likely to benefit humanity as a whole, unconstrained by a need to generate financial return. Since our research is free from financial obligations, we can better focus on a positive human impact.
Whether the case is any good is another matter.
Try to contribute to the conversation, though.
What you say is also untrue, there's a minimum set of requirements that have to be met regarding discovery, etc.
I can’t tell if Elon is saying that the tweet is misinformed or OpenAI’s behavior is reprehensible
Foundations and charitable organizations that pubically get their funding are a different story but I'm talking about non profit companies.
I even had one fellow say that the green bay packers were less corrupt than the other for profit nfl teams , which sounds ridiculous.
The NFL’s non-profit status is a farce though. Similarly, their misuse of copyright (“you cannot discuss this broadcast”) and the trademark “Super Bowl” (“cannot be used in factual statements regarding the actual Super Bowl”) should have their ownership of that ip revoked, if only because it causes massive confusion about the underlying law with a big chunk of the US population.
This claim he made was likely helpful in ensuring the OpenAI team’s willingness to bring him back after he was temporarily ousted by the board last year for alleged governance issues. (Basically: “don’t worry about me guys, I’m in this for the mission, not personal enrichment”)
Since his claim likely helped him get re-hired, he can’t claim it was immaterial.
I really hope someone from the SEC scrutinizes him someday. The Singularity is too important to let it be run by someone with questionable ethics.
https://www.sec.gov/resources-small-businesses/capital-raisi...
It is quite likely they knew the latter as relatively low risk expected outcome .
Even at 100M exit, which by valley standards (even in 2010s) is not a lot, 1-2% (after further rounds of dilution) would have yielded 1-2M return . A 200x return for very little downside i.e. a gift .
There is a reason why there is FOMO and little due diligence for really hot startups amongst VCs , most times it is about access to the round which is difficult rather than risk of returns, we only read about the spectacular failures like FTX . We don’t hear about the Stripe, AirBnb, or Figma, OpenAI or spaceX funding rounds .
I guess being michael jordan of listening does't help with imagination
I was working at dunkin donuts when i was a teenager.
20 yr age gap in a relationship is still frowned upon in our society unfortunately.
Also I guess you were tracking sam's legal age closer than Sam himself because he mentioned that he was a teenager when he met PG in an interview. SV is creepier than anyone could ever imagine.
I'm not sure how it's relevant that you worked in a donut shop? Surely you're aware that isn't the peak of over-achievement at the age of 20? God, watch the Olympics if you want to see a bunch of very determined young people. Or go to an Open Source conference. Or, as it were, YC. Sam was only slightly younger than normal there. Some 20 year olds have accomplished a lot.
You seem to really want to create a villain out of the situation, which is kind of weird. Mentors and investors are usually older, and 20 years isn't rare. That's hardly Silicon Valley specific.
You also repeated the false thing about him getting a stake in Stripe as a teenager again. Again, Stripe came into existence when Sam was 25. There's just no version of that story that works. I don't know about Sam's Stripe investment, but at that point he was already around YC a lot, even though Loopt was still going. He probably just got in with other angel investors. But at that point he was in his mid-20s and running a Sequoia-backed company, so that's not especially weird.
(There's genuine stuff to be critical of in the trajectory of OpenAI, but this seems like a really weird spot to latch onto.)
PG:
"Sam Altman, the co-founder of Loopt, had just finished his sophomore year when we funded them, and Loopt is probably the most promising of all the startups we've funded so far. But Sam Altman is a very unusual guy. Within about three minutes of meeting him, I remember thinking 'Ah, so this is what Bill Gates must have been like when he was 19.'"
Are you being sarcastic by comparing sam with olympians?
What exactly did Sam accomplish when he met PG to declared as "bill gates" or to get 50 million for his startup within 15 minutes of meeting PG ?
He was practicing being "Michael jordan of listening" ( another PG quote) since he was 5 like olympians ?
I was in YC a few batches later and met Paul Graham and Sam in that era. I remember walking around San Fracisco with Sam and him telling me about Loopt. He was a few years younger than me (I was 29, he was 24), and I remember being impressed by him.
And it's possible that Sam listed his age at 19 on his YC application, and that's what PG was going on. He would have probably still been 19 when he filled out the application. Again, this isn't hard to verify -- his birthday is on Wikipedia, and he was in the summer batch of 2005. Interviews are about a month before the batch starts. But there's not really a lot of my point that hinges on if it was a month before or a month after his birthday when they met. More my point was that the stuff about him investing in Stripe as a teenager because PG "gave" it to him is completely bogus.
It really seems like you have an axe to grind here, and I'm not completely sure why. Again, I think some of the stuff that's happened later in OpenAI is worthy of criticism, but that doesn't mean you have to reinterpret everything that happened before that through some bogyman lens.
Yea because ppl getting unfair leg up because they were chosen as the "next bill gates" by a SV white male because they look like them is merely an "axe to grind".
You still haven't answered why you think he is like an olympian when he met PG other than "He is impressive because he is impressive".
I feel like i am in some kind of weirdo land here with totally ridiculous boasts about someone that no one can name an actual accomplishment
> like an olympian
> michaal jordan of listening
> bill gates at 19
> his brain will be cloned by 2029
You guys need to send this to HBO for next Silicon Valley season.
The whole "white man" thing is also a complete straw man. The other two YC-founders-turned YC CEOs of that era were Michael Seibel and Garry Tan, neither of whom are white.
It sounds like what you're offended by is the whole YC process -- that there are quick interviews that (back then) translated to small amounts of funding -- that literally the decision was made in a single interview. But that wasn't anything specific to Sam; that's how it worked for everyone. You can find that stupid if you want to, but then might I suggest this is an odd forum to hang out on if you find that to be offensive?
Exactly. I was comparing myself to him when i mentioned that I worked at a donut shop. At that age I ( and many others) were indistinguishable from him. I went to an ivy league too btw.
Your olympian thing is absurd here because a teenager destined to be an olympian is indeed very distinguishable from his/her peers very easily, ppl can tell why this person is special.
You keep saying Sam was special ( next bill gates) but fail to tell me why. I asked you multiple times too , but instead you keep attacking me instead for not accepting circular "he is impressive because he is impressive" .
> But that wasn't anything specific to Sam; that's how it worked for everyone.
I just told you but you keep ignoring. Did PG mention anyone else was bill gates or Michael Jordan to his VC friends and publicly in interviews. (Ironic, given Michael Jordan is one the most impressive athletes that came from no where ). Or continuously give him leg up despite failed ventures ( loopt or whatever). I cannot think of anyone else who failed upwards like Sam because he had PG to back him with ridiculous and vacuous pumping of Sam's so called genius.
Oh yea you would rather run way and feel smug about some pedantic age thing than substantiate why you think same was "impressive" when he met 20s.
Is it well played if you simply decide to lie brazenly? Anyone can win at monopoly if they decide to steal from the bank.
And, most importantly: non-profit charities (not the only kind of nonprofit, but presumably what OpenAI was) are legally obligated to operate “for the public good”. That’s why they’re tax exempt: the government is basically donating to them, with the understanding that they’re benefiting the public indirectly by doing so, not just making a few people rich.
In my understanding, this is just blatant outright fraud that any sane society would forbid. If you want to start a for-profit that’s fine, but you’d have to give away the nonprofit and its assets, not just roll it over to your own pocketbook.
God I hope Merrick Garland isn’t asleep at the wheel. They’ve been trust busting like mad during this administration, so hopefully they’re taking aim at this windmill, too.
Little chance of that as Sama is a big time Democrat fundraiser and donor.
Can’t find a good source for both rn but this one has alphabet in the top 50 nationwide for this election: https://www.opensecrets.org/elections-overview/top-organizat...
edit: and Sam Altman isn’t exactly donating game changing amounts — around $300K in 2020, and seemingly effectively nothing for this election. That’s certainly nothing to sneeze at as an individual politician, but that’s about 0.01% of his net worth (going off Wikipedia’s estimate of $2.8B, not counting the ~$7B of OpenAI stock coming his way).
https://www.dailydot.com/debug/openai-sam-altman-political-d...
When you see any numbers for corporations contributing to political campaigns, that's actually just measuring the contributions from the employees of those corporations. That's why most corporations "donate to both parties"--because they employ both Republicans and Democrats.
Name names. We can look it up.
Again, I am not a lawyer but that makes no sense. Otherwise, anyone can claim the non-profit? So clearly there are some beneficial owners out there somehow.
1: In principle; in practice, well, we'll see with this one!
In practice it’s doable though. You can just create a new legal entity and move stuff and/or do future value creating activity in the new co. IF everyone is on board with the plan on both sides of the move then that’s totally doable with enough lawyers and accountants
I'm wondering if OpenAI's charter might provide a useful legal angle. The charter states:
>OpenAI’s mission is to ensure that [AGI ...] benefits all of humanity.
>...
>We commit to use any influence we obtain over AGI’s deployment to ensure it is used for the benefit of all, and to avoid enabling uses of AI or AGI that harm humanity or unduly concentrate power.
>Our primary fiduciary duty is to humanity. We anticipate needing to marshal substantial resources to fulfill our mission, but will always diligently act to minimize conflicts of interest among our employees and stakeholders that could compromise broad benefit.
>...
>We are committed to doing the research required to make AGI safe, and to driving the broad adoption of such research across the AI community.
>We are concerned about late-stage AGI development becoming a competitive race without time for adequate safety precautions. [...]
>...
I'm no expert here, but to me, this charter doesn't appear to characterize OpenAI's behavior as of the year 2024. Safety people have left, Sam has inexplicably stopped discussing risks, and OpenAI seems to be focused on racing with competitors. My question: Is the charter legally enforceable? And if so, could it make sense for someone to file an additional lawsuit? Or shall we just wait and see how the Musk lawsuit plays out, for now?
I'm curious about the "fiduciary duty" part. As a member of humanity, it would appear that OpenAI has a fiduciary duty to me. Does that give me standing? Suppose I say that OpenAI compromises my safety (and thus finances) by failing to discuss risks, having a poor safety culture (as illustrated by employee exits), and racing. Would that fly?
Even if you donated to them, all states I know of assign sole oversight for proper management of those funds to the state AG. If you donate to a food bank and they use the money to buy personal Ferraris instead of helping the hungry, that's clearly illegal, but you'd be out the money either way, so you wouldn't have standing to sue. The attorney general has to sue for mismanagement of funds. If you feel OpenAI is violating their charter, I would definitely encourage writing to Mrs. Jennings to voice that opinion.
I don't know the laws of Delaware well, but I would be surprised if he has standing even as a donor.
The IRS isn’t stupid. The rules on what counts as taxable income and what the nonprofit can take tax-free have been around for decades.
https://www.propublica.org/article/how-the-irs-was-gutted (2018)
> An eight-year campaign to slash the agency’s budget has left it understaffed, hamstrung and operating with archaic equipment. The result: billions less to fund the government. That’s good news for corporations and the wealthy.
You sign an exclusive, non-revocable licensing agreement. Ownership of the original IP remains 100% with the original startup.
Now, this only works if the non-profit's board is on-board.
I think the real issue Musk was complaining about is that sama is quickly becoming very wealthy and powerful and Musk doesn't want any competition in this space.
Hopefully some people watching all this realize that the people running many of these big AI related projects don't care about AI. Sam Altman is selling a dream about AGI to help make himself both wealthier and more powerful, Elon Musk is doing the same with electric cars or better AI.
People on HN are sincerely invested in the ideas behind these things, but it's important to recognize that the people pulling the strings largely don't care outside how it benefits them. Just one of the many reasons, at least in AI, truly open source efforts are essential for any real progress in the long run.
There's a lot of jurisdiction around preventing this sort of abuse of the non-profit concept.
The reason why the people involved are not on trial right now is a bit of a mystery to me, but could be a combination of:
* Still too soon, all of this really took shape in the past year or two.
* Only Musk has sued them, so far, and that happened last month.
* There's some favoritism from the government to the leading AI company in the world.
* There's some favoritism from the government to a big company from YC and Sam Altman.
I do believe Musk's lawsuit will go through. The last two points are worth less and less with time as AI is being commoditized. Dismantling OpenAI is actually a business strategy for many other players now. This is not good for OpenAI.
Which ones exactly?
NVIDIA is drinking sweet money from OpenAI.
Microsoft & Apple are in cahoots with it.
Meta/Facebook seems happy to compete with OpenAI on a fair playing field.
Anthropic lacks the resources.
Amazon doesn't seem to care.
Google is asleep.
NVIDIA makes money from any company doing AI. I would be surprised if OpenAI was a whole digit percentage of their revenue.
>Microsoft & Apple are in cahoots with it.
Nope. Apple is using OpenAI to fill holes their current model is not good at. This doesn't sound like a long-term partnership.
>Meta/Facebook seems happy to compete with OpenAI on a fair playing field.
They want open source models to rule, obliterating proprietary models out of existence, while at it.
>Anthropic lacks the resources.
Hence why it would be better for them if OpenAI would not exist. It's the same with all other AI companies out there.
>Amazon doesn't seem to care.
Citation needed, AWS keeps putting out products which are their market leaders, they just don't make a big fuzz about it.
>Google is asleep.
I'll give you this one. I have no idea why they keep Pichai around.
As opposed to? The euphemism "I wouldn't be surprised" usually means you think what you're saying. If you negate that you're saying what you _don't_ think is the case? I may be reading too much into whats probably a typo.
It is not publicly known how much revenue Nvidia gets from OpenAI, but it is likely more than 1%, and they may be one of the top 4 unnamed customers in their 10Q filing, which would mean at least 10% and $3 billion [0].
That's not nothing.
[0] https://www.yahoo.com/tech/nvidia-gets-almost-half-revenue-0...
OpenAI is miles ahead in terms of ecosystem and platform integration. Google can come up with long context windows and cool demos all they want, OpenAI built a lot of moat while they were busy culling products :)
I didn't realise it was that bad.
Amazon gets paid either way, because even if open ai doesn’t use them, where are you going to cloud your api that’s talking with open ai?
If open ai looks weakened I think we’ll see everyone else has a service they want you to try. But there’s no use in making much noise about that, especially during an election year. No matter who wins, all the rejected everywhere will blame AI, and who knows what that will look like. So, sit back and wait for the leader of the pack to absorb all the damage.
I feel like this is quite a slippery slope, though. Should we also give small companies a right to violate trademarks? Copyright? Kill people? These could also give them a chance to compete against big players.
As OpenAI found its product-market fit, the early visionaries are not needed anymore (although I'm sure the people working there are still amazing)
most would happily sell their soul and deal with any mess to reach $150B valuation
Obviously we all think of the NFL as a big money organisation, but it basically just organises the fixtures and the referees. The teams make all the money.
They pay dividends to the teams, however, yes. But all that revenue (which is distinct from team revenue) is actually legally earned by the NFL itself.
Interesting timing of the news since Murati left today, gdb is 'inactive' and Sutskevar has left to start his own company. Also seeing few OpenAI folks announcing their future plans today on X/Twitter
Although I guess it doesn't really matter. What if we all understood climate change earlier? wouldn't really have made a difference anyway
Altman was fucking with OpenAI for long before the board left in protest, since about the time Elon Musk had to leave due to Tesla's AI posing a conflict of interest. He got more and more brazen with the whole fake-altruism shit, up to and including contradicting every point in their mission statement and promise to investors in the "charity."
>Beethoven's reaction to Napoleon Bonaparte's declaration of himself as Emperor of France in May 1804 was to violently tear Napoleon's name out of the title page of his symphony, Bonaparte, and rename it Sinfonia Eroica
>Beethoven was furious and exclaimed that Napoleon was "a common mortal" who would "become a tyrant"
Sketchy.
This whole silicon valley attitude of fake effective altruism, "I do it for the good of humanity, not for the money (but I actually want a lot of money)" fake bullshit is so transparent and off-putting.
@sama, for the record - I am not saying making is a bad thing. Labor and talent markets should be efficient. But when you pretend to be altruistic when you are obviously not, then you come off hypocritical instead of altruistic. Sell out.
But buying a $4M car while saying you do not car about money is a mis-alignment between words and actions, which comes off untrustworthy.
Like you might say “I don’t really need another car we already have 8 at home between the two of us driving so why would I need a 9th?”
I guess technically it's supposed to play some role in making sure OpenAI "benefits humanity". But as we've seen multiple times, whenever that goal clashes with the interests of investors, the latter wins out.
That entity will scrape the internet and train the models and claim that "it's just research" to be able to claim that all is fair-use.
At this point it's not even funny anymore.
It does suit the modus operandi of a number of American companies that start out as literally illegal/criminal operations until they get big and rich enough to pay a fine for their youthful misdeeds.
By the time some of them get huge, they're in bed with the government to dominate the market.
Or we're all talking about and envisioning some specific little subset of artists. I suspect you're trying to pretend that someone with a literal set of paintbrushes living in a shitty loft is somehow having their original artwork stolen by AI despite no high resolution photography of it existing on the internet. I'm not falling for that. Be more specific about which artists are losing their livelihoods.
If we take your argument to it's logical conclusion, all progress is inherently bad, and should be stopped.
I deposit instead that the real problem is that we tied people's ability to afford basic necessities to how much output they can produce as a cog in our societal machine.
Yes, because if you depend on some overarching organisation or person to give it to you, you are fucked 100% of the time due this dependency.
If AI replaces millions of jobs, it will be a net negative in job availability for working class people.
I agree with your last point, the way the system is set up is incompatible with the looming future.
There is no similar net creation of jobs for society if jobs are eliminated by AI, and it's even worse than that because many of the jobs are specialized, high-skill positions that can't be transferred to other careers easily. It goes without saying that it also includes millions of low-skill jobs like cashiers, stockers, data entry, CS reps, etc. Generally people who are already struggling to get enough hours and feed their families as it is.
Recording devices permitted artists to sell more art.
Many of the uses of AI people get most excited about seem to be cutting the expensive human creators out of the equation.
Current AI can greatly elevate what a beginning artist can produce. If you have a decent grasp of proportions, perspective and good ideas, but aren't great at drawing, then using AI can be a huge quality improvement.
On the other hand if you're a top expert that draws quickly and efficiently it's quite possible that AI can't do very much for you in a lot of cases, at least not without a lot of hand tuning like training it on your own work first.
So yeah it had a profound effect, but we got consent for the parts that fundamentally relied on other people.
It's what everyone else does. The entitlement has to stop.
You're advocating for destroying all AI or ensuring a monopoly by corporations. Whose side are you actually on?
Irrelevant. The law does not care about feasibility of breaking it.
If I decide to run a hit man business, that's also infeasible. Dealing with the arrests and fines would be too much. The conclusion then is not to bend the law to make murder legal. The conclusion is my business is illegitimate, and it's the civic duty of my Country to make sure it fails.
> Whose side are you actually on?
The people making the content that corps are profiting big off of. They should pay a license.
Nope. The law will side with whoever pays the most. Once OpenAI solidifies its top position, only then will regulations kick in. Take YouTube, for example—it grew thanks to piracy. Now, as the leader, ContentID and DMCA rules work in its favor, blocking competition. If TikTok wasn’t a copyright-ignoring Chinese company, it would’ve been dead on arrival.
If it's only OK to scrape, lossy-compress, and redistribute book-paragraphs when it gets blended into a huge library of other attempts, then that's only going to empower big players that can afford to operate at that scale.
I've no idea if it could be valid when it comes to OpenAI, but it does seem to be a general concept designed to counter wrongdoers who take a little value from a lot of people?
The crazy thing is that there hasn't been an injunction to make them stop.
burning the bridge so nobody else can legally scrape, that's the line.
Assets like the Internet Archive, though, should be protected at all costs.
Update: ML doesn't copy information. It can merely memorise some small portions of it.
https://www.copyright.gov/title37/201/37cfr201-14.html
§ 201.14 Warnings of copyright for use by certain libraries and archives.
....
The copyright law of the United States (title 17, United States Code) governs the making of photocopies or other reproductions of copyrighted material.
Under certain conditions specified in the law, libraries and archives are authorized to furnish a photocopy or other reproduction. One of these specific conditions is that the photocopy or reproduction is not to be “used for any purpose other than private study, scholarship, or research.” If a user makes a request for, or later uses, a photocopy or reproduction for purposes in excess of “fair use,” that user may be liable for copyright infringement.
This institution reserves the right to refuse to accept a copying order if, in its judgment, fulfillment of the order would involve violation of copyright law.
It would be perfectly legal for me to go to the library and make photocopies of works. I could even take them home and use the photocopies as reference works write an essay and publish that. If {random person} took my photocopied pages and then sold them, that would likely go beyond the limits placed for how the photocopied works from the library may be used.
A more fitting metaphor would be something like... If you had the ability to read all the books in the library extremely quickly, and to make useful mental connections between the information you read such that people would come to you for your vast knowledge, should you be allowed in the library?
The anti-AI stance is what is baffling to me. The path trotten is what got us here and obviously nobody could have paid people upfront for the wild experimentation that was necessary. The only alternative is not having done it.
Given the path it has put as in, people either are insanely cruel or just completely detached from reality when it comes to what is necessary to do entirely new things.
"Hugely beneficial" is a stretch at this point. It has the potential to be hugely beneficial, sure, but it also has the potential to be ruinous.
We're already seeing GenAI being used to create disinformation at scale. That alone makes the potential for this being a net-negative very high.
Perhaps the biggest “needs citation” statement of our time.
Not in any weirdly-self-aggrandizing "our tech is so powerful that robots will take over" sense, just the depressingly regular one of "lots of people getting hurt by a short-term profitable product/process which was actually quite flawed."
P.S.: For example, imagine having applications for jobs and loans rejected because all the companies' internal LLM tooling is secretly racist against subtle grammar-traces in your writing or social-media profile. [0]
To continue one of the analogies: Plenty of people and industries legitimately benefited from the safety and cost-savings of asbestos insulation too, at least in the short run. Even today there are cases where one could argue it's still the best material for the job--if constructed and handled correctly. (Ditto for ozone-destroying chlorofluorocarbons.)
However over the decades its production and use grew to be over/mis-used in so very many ways, including--very ironically--respirators and masks that the user would put on their face and breathe through.
I'm not arguing LLMs have no reasonable uses, but rather that there are a lot of very tempting ways for institutions to slot them in which will cause chronic and subtle problems, especially when they are being marketed as a panacea.
We have a term for that, it's called "luddite". Those were english weavers who would break in to textile factories and destroy weaving machines at the beginning of the 1800s. With the extreme rare exception, all cloth is woven by machines now. The only hand made textiles in modern society are exceptionally fancy rugs, and knit scarves from grandma. All the clothing you're wearing now are woven by a machine, and nobody gives this a second thought today.
https://en.wikipedia.org/wiki/I%27m_alright,_Jack
Except, we are all Jack.
Maybe it would have been better for humanity if the Luddites won.
It is not possible to rehabilitate the Luddites. If you insist on attempting to do so, there are better venues.
No, that's apples-to-oranges. The goals and complaints of Luddites largely concerned "who profits", the use of bargaining power (sometimes illicit), and economic arrangements in general.
They were not opposing the mechanization by claiming that machines were defective or were creating textiles which had inherent risks to the wearers.
I have never thought of being anti-AI as “Luddite”, but actually this very description of “Luddite” does sound like the concerns are in fact not completely different.
Observe:
Complaints about who profits? Check; OpenAI is earning money off of the backs of artists, authors, and other creatives. The AI was trained on the works of millions(?) of people that don’t get a single dime of the profits of OpenAI, without any input from those authors on whether that was ok.
Bargaining power? Check; OpenAI is hard at work lobbying to ensure that legislation regarding AI will benefit OpenAI, rather than work against the interests of OpenAI. The artists have no money nor time nor influence, nor anyone to speak on behalf of them, that will have any meaningful effect on AI policies and legislation.
Economic arrangements in general? Largely the same as the first point I guess. Those whose works the AI was trained on have no influence over the economic arrangements, and OpenAI is not about to pay them anything out of the goodness of their heart.
The Luddites were actually a fascinating group! It is a common misconception that they were against technology itself, in fact your own link does not say as much, the idea of “luddite” being anti-technology only appears in the description of the modern usage of the word.
Here is a quote from the Smithsonian[1] on them
>Despite their modern reputation, the original Luddites were neither opposed to technology nor inept at using it. Many were highly skilled machine operators in the textile industry. Nor was the technology they attacked particularly new. Moreover, the idea of smashing machines as a form of industrial protest did not begin or end with them.
I would also recommend the book Blood in the Machine[2] by Brian Merchant for an exploration of how understanding the Luddites now can be of present value
1 https://www.smithsonianmag.com/history/what-the-luddites-rea...
2 https://www.goodreads.com/book/show/59801798-blood-in-the-ma...
They had very rational reasons for trying to slow the introduction of a technology that was, during a period of economic downturn, destroying a source of income for huge swathes of working class people, leaving many of them in abject poverty. The beneficiaries of the technological change were primarily the holders of capital, with society at large getting some small benefit from cheaper textiles and the working classes experiencing a net loss.
If the impact of LLMs reaches a similar scale relative to today's economy, then it would be reasonable to expect to see similar patterns - unrest from those who find themselves unable to eat during the transition to the new technology, but them ultimately losing the battle and more profit flowing towards those holding the capital.
> "our tech is so powerful that robots will take over"
> "lots of people getting hurt by a short-term profitable product/process which was actually quite flawed."
You response assumes the former, but it's my understanding the Luddite's actual position was the latter.
> Luddites objected primarily to the rising popularity of automated textile equipment, threatening the jobs and livelihoods of skilled workers as this technology allowed them to be replaced by cheaper and less skilled workers.
In this sense, "Luddite" feels quite accurate today.
We don't have to imagine such things, really, as that's extremely common with humans. I would argue that fixing such flaws in LLMs is a lot easier than fixing it in humans.
I currently work in the HR-tech space, so suppose someone has a not-too-crazy proposal of using an LLM to reword cover-letters to reduce potential bias in hiring. The issue is that the LLM will impart its own spin(s) on things, even when a human would say two inputs are functionally identical. As a very hypothetical example, suppose one candidate always does stuff like writing out the Latin like Juris Doctor instead of acronyms like JD, and then that causes the model to end up on "extremely qualified at" instead of "very qualified at"
The issue of deliberate attempts to corrupt the LLM with prompt-injection or poisonous training data are a whole 'nother can of minefield whack-a-moles. (OK, yeah, too far there.)
I just don't think your original comment was entirely fair. IMO, LLMs and related technology will be looked at similarly as the Internet - certainly it has been used for bad, but I think the good far outweighs the bad, and I think we have (and continue to) learn to deal with the issues with it, just as we will with LLMs and AI.
(FWIW, I'm not trying to ignore the ways this technology will be abused, or advocate for the crazy capitalistic tendency of shoving LLMs in everything. I just think the potential for good here is huge, and we should be just as aware of that as the issues)
(Also FWIW, I appreciate your entirely reasonable comment. There's far too many extreme opinions on this topic from all sides.)
Some not-problems, presented as though they are:
"How can we prevent the untimely eradication of Polio?"
"How can we prevent bot network operators from being unfairly excluded from online political discussions?"
"How can we enable context-and-content-unaware text generation mechanisms to propagate throughout society?"
For example, MKUltra tried to solve a problem: "How can I manipulate my fellow man?" That problem still exists today, and you bet AI is being employed to try to solve it.
History is littered with problems such as these.
Yes, we are clearly talking about things to mostly still come here. But if you assign a 0 until its a 1 you are just signing out of advancing anything that's remotely interesting.
If you are able to see a path to 1 on AI, at this point, then I don't know how you would justify not giving it our all. If you see a path and in the end using all of human knowledge up to this point was needed to make AI work for us, we must do that. What could possibly be more beneficial to us?
This is regardless of all issues the will have to be solved and the enormous amount of societal responsibility this puts on AI makers — which I, as a voter, will absolutely hold them accountable for (even though I am actually fairly optimistic they all feel the responsibility and are somewhat spooked by it too).
But that does not mean I think it's responsible to try and stop them at this point — which the copyright debate absolutely does. It would simply shut down 95% of AI, tomorrow, without any other viable alternative around. I don't understand how that is a serious option for anyone who roots for us.
Oh, the humanity! Who will write our third-rate erotica and Russian misinformation in a post-AI world?
I don’t think that the consumer LLMs that openai is pioneering is what need optimism.
AlphaFold and other uses of the fundamental technology behind LLMs need hype.
Not OpenAI
AlphaFold is a game changer for medical R&D. Everyone should be hyped for that.
They also are leveraging these same ML techniques for detecting kelp forest off the coast of Australia for preservation.
Alphabet isn’t a great company, but that does not mean the good they do should be ignored.
Much more deserving than chatgpt. Productifyed LLMs are just an attempt to make a new consumer product category.
I think you raise some interesting concerns in your last paragraph.
> enormous amount of societal responsibility this puts on AI makers — which I, as a voter, will absolutely hold them accountable for
I'm unsure of what mechanism voters have to hold private companies accountable. Fir example, whenever YouTube uses my location without me ever consenting to it - where is the vote to hold them accountable? Or when Facebook facilitates micro targeting of disinformation - where is the vote? Same for anything AI. I believe any legislative proposals (with input from large companies) is very likely more to create a walled garden than to actually reduce harm.
I suppose no need to respond, my main point is I don't think there is any accountability thru the ballot when it comes to AI and most things high-tech.
Firstly, *skeptics.
Secondly, being skeptical doesn't mean you have no optimism whatsoever, it's about hedging your optimism (or pessimism for that matter) based on what is understood, even about a not-fully-understood thing at the time you're being skeptical. You can be as optimistic as you want about getting data off of a hard drive that was melted in a fire, that doesn't mean you're going to do it. And a skeptic might rightfully point out that with the drive platters melted together, data recovery is pretty unlikely. Not impossible, but really unlikely.
Thirdly, OpenAI's efforts thus far are highly optimistic to call a path to true AI. What are you basing that on? Because I have not a deep but a passing understanding of the underlying technology of LLMs, and as such, I can assure you that I do not see any path from ChatGPT to Skynet. None whatsoever. Does that mean LLMs are useless or bad? Of course not, and I sleep better too knowing that LLM is not AI and is therefore not an existential threat to humanity, no matter what Sam Altman wants to blither on about.
And fourthly, "wanting" to stop them isn't the issue. If they broke the law, they should be stopped, simple as. If you can't innovate without trampling the rights of others then your innovation has to take a back seat to the functioning of our society, tough shit.
> The anti-AI stance is what is baffling to me
I don't see s lot of anti AI but instead I see a concern for how it's just being managed and controlled by the larger companies with resources that no start up could dream. Open AI was to release it's models and be well.. Open but fine they're not. But their behaviour of how things are proceeding are questionable and unnecessarily aggravating.
I don't think this is the "ends justify the means" argument you think it is.
Automobiles allow people to travel great distances over short periods of time, increase physical work capacity, allow for building massive structures, and allow for farming insane amounts of food.
Both the internet and automobiles have positively affected my life, and I assume the lives of many others. How are any of these aimless questions?
And who is the one calling for action?
Sorry for being dense, but I'm trying to understand if I'm the "strong" or the "weak" in your analogy.
The work of artists, authors, etc.
I know currently the legal situation is messy, but that's exactly the point, anyone who can't engage in lengthy legal battle and defend their position in court are being sacrificed. The companies behind LLMs are spending hundreds of millions of dollars in lobbying and exploiting loopholes.
Let's be real without the data there wouldn't be LLMs, so it crazy that some people are downplaying its significance or value, while on the other hand they're losing sleep over finding fresh sources to scrape.
The big publishers seem to have given up and decided it's best to reach agreement with their counterparts, while independent authors are given the finger.
> The work of artists, authors, etc.
Define "a lot"? Most people barely know how to use their email. Even among the minority who do actively use "AI" and excited about it, outside of ML engineers they aren't well-informed or aware what data is used for training, or even what training means and how these models work to begin with.
> People are excited and enthusiastic about AI and are actively reaping the benefits of progress.
Except the terms were already violated in the initial training phase before the services were even public and saw adoption. That's like pointing at a rape victim who got some form of compensation later, saying:
see how she's "reaping the benefits"
By the time some people started raising concerns, OpenAI claimed the cat was already out of the bag and "if we didn't do it, someone else will, so deal with it."
Similar to privacy, just because some people don't care, lack awareness, or don't want the hassle of fighting for it, doesn't justify taking it away from others.
This is repackaging content, laundering it, and reselling it.
As others have noted, IP law has lots of problems; Sam Altman et al are exploiting the gap left between the speed of technology and law and using their own version of social good without waiting for the consent of those they're exploiting.
The anti-AI stance is what is baffling to me.
And isn’t this what basically every single scholar in the field says they want, anyway - safe, intentional, controlled deployment?
As you can tell from the above, I’m as far from being “anti-AI” or technically pessimistic as one can be — I plan to dedicate my life to its safe development. So there’s at least one counterexample for you to consider :)
OpenAI's case is especially egregious, with the entire starting as 'open' and reaping the benefits, then doing its best in every way to shut the door after itself by scaring people over AI apocalypses. If your argument is seriously that it is necessary to shamelessly steal and lie to do new things, I question your ethical standards, especially in the face of all the openly developed models out there.
I do not have confidence in the Supreme Court in general, and I think there's a real risk that in deciding on AI training they upend copyright of digital materials in a way that makes it worse for everyone.
It's completely unprecedented.
We allowed scraping images and text en masse when search engines used the data to let us find stuff.
We allow copying of style, and don't allow writing styles and aesthetics to be copyrighted or trademarked.
Then AI shows up, and people change lanes because they don't like the results.
One of the things that made me tilt towards the side of fair use was a breakdown of the Stable Diffusion model. The SD2.1 base model was trained on 5.85 billion images, all normalized to 512x512 BMP. That's 1MB per images, for a total of 5.85PB of BMP files. The resulting model is only 5.2GB. That's more than 99.999999% data loss from the source data to the trained set.
For every 1MB BMP file in the training dataset, less than 1byte makes it into the model.
I find it extremely difficult to call this redistribution of copyrighted data. It falls cleanly into fair use.
Their arguments against this amounts to "we're not using it like they intend it to be used, so it's fine if we obtain it illegally", and that's a bs standard, totally divorced from any legal reality.
Fair Use covers certain transformative uses, certainly, but it doesn't cover illegal obtaining of the content.
You can't pirate a book just because you want to use it transformatively (which is exactly what they've done), and that argument would never hold up for us as individuals, so we sure as hell shouldn't let tech companies get a special carve-out for it.
I'm surprised people are surprised.
>> That entity will scrape the internet and train the models and claim that "it's just research" to be able to claim that all is fair-use.
a lot of people and entities do this though... openAI is in the spotlight, but scraping everything and selling it is the business model for a lot of companies...
In my eyes, all genAI companies/tools are the same. I dislike all equally, and I use none of them.
That's the business model of lots of companies. Take, collect and collate data, put it in a new format more useful for your field/customers, resell.
I'm ignorant on this topic so please excuse me. Why did `AI` happen now? What was the secret sauce that OpenAI did that seemed to make this explode into being all of a sudden?
My general impression was that the concept of 'how it works' existed for a long time, it was only recently that video cards had enough VRAM to hold the matrix(?) within memory to do the necessary calculations.
If anybody knows, not just the person I replied to.w.r.t. Branding.
AI has been happening "forever". While "machine learning" or "genetic algorithms" were more of the rage pre-LLMs that doesn't mean people weren't using them. It's just Google Search didn't brand their search engine as "powered by ML". AI is everywhere now because everything already used AI and now the products as "Spellcheck With AI" instead of just "Spellcheck".
w.r.t. Willingness
Chatbots aren't new. You might remember Tay (2016) [1], Microsoft's twitter chat bot. It should seem really strange as well that right after OpenAI releases ChatGPT, Google releases Gemini. The transformers architecture for LLMs is from 2014, nobody was willing to be the first chatbot again until OpenAI did it but they all internally were working on them. ChatGPT is Nov 2022 [2], Blake Lemoine's firing was June 2022 [3].
[1]: https://en.wikipedia.org/wiki/Tay_(chatbot)
[2]: https://en.wikipedia.org/wiki/ChatGPT
[3]: https://www.npr.org/2022/06/16/1105552435/google-ai-sentient
Thanks for the links too!
1986: Geoffrey Hinton publishes the backpropagation algorithm as applied to neural networks, allowing more efficient training.
2011: Jeff Dean starts Google Brain.
2012: Ilya Sutskever and Geoffrey Hinton publish AlexNet, which demonstrates that using GPUs yields quicker training on deep networks, surpassing non-neural-network participants by a wide margin on an image categorization competition.
2013: Geoffrey Hinton sells his team to the highest bidder. Google Brain wins the bid.
2015: Ilya Sutskever founds OpenAI.
2017: Google Brain publishes the first Transformer, showing impressive performance on language translation.
2018: OpenAI publishes GPT, showing that next-token prediction can solve many language benchmarks at once using Transformers, hinting at foundation models. They later scale it and show increasing performance.
The reality is that the ideas for this could have been combined earlier than they did (and plausibly future ideas could have been found today), but research takes time, and researchers tend to focus on one approach and assume that another has already been explored and doesn’t scale to SOTA (as many did for neural networks). First mover advantage, when finding a workable solution, is strong, and benefited OpenAI.
Honestly, those are not the missing parts that most matter IMO. The evolution of the concept of attention across many academic papers which fed to the Transformer is the big missing element in this timeline.
Not really:
History: https://arxiv.org/abs/2212.11279 (75 pp.)
Survey: https://arxiv.org/abs/1404.7828 (88 pp.)
Conveniently skim-read over the course of the four weekends on one month.
But out of the co-founders, especially if we believe Elon's and Hinton's description of him, he may have been the one that mattered most for their scientific achievements.
We've had upgrades to hardware, mostly led by NVidia, that made it possible.
New LLMs don't even rely that much on that aforementioned older architecture, right now it's mostly about compute and the quality of data.
I remember seeing some graphs that shows that the whole "learning" phenomena that we see with neural nets is mostly about compute and quality of data, the model and optimizations just being the cherry on the cake.
Don’t they all indicate being based on the transformer architecture?
> not entirely because of transformers but because of the hardware
Kaplan et al. 2020[0] (figure 7, §3.2.1) shows that LSTMs, the leading language architecture prior to transformers, scaled worse because they plateau’ed quickly with larger context.
As a moral fig leaf. They can always point to it when the press calls -- "see it is a non-profit".
Low level employees are there for the money, not for the drama.
the well known scammer successfully scammed everyone twice. obviously he's keeping it around for the third (and forth...) time
A tale as old as time. Some of us could see it, from afar <says while scratching gray, dusty beard>. Lack of upvotes and excitement does not mean support, but how to account for that in these times? <goes away>
Going for-Profit, and several top exec leaving at same time? Before getting the money?
"""Question: why would key people leave an organization right before it was just about to develop AGI?" asked xAI developer Benjamin De Kraker in a post on X just after Murati's announcement. "This is kind of like quitting NASA months before the moon landing," he wrote in a reply. "Wouldn't you wanna stick around and be part of it?"""
https://arstechnica.com/information-technology/2024/09/opena...
Is this the beginning of the end for OpenAI?
Also because you know… the non-profit tried to strangle the for profit to take over control when they tried to oust Sam, so there’s that.
You can name your company "ThisProductWillCureYouFromCancer" and the FDA cannot do a thing about it if you put it on a bottle of herbal pills.
- Technically true means you will probably win any lawsuit they bring
- In practice means that they will in fact bring a lot of lawsuits, making it very expensive for you and difficult for you to operate. They will probably find excuses to harass you over every little thing, they will harass you over lots of details that are technically required but rarely enforced in practice. You'll constantly be getting inspected and audited, they will bring lawsuits for other, apparently unrelated things.
I wonder though whether Microsoft is still interested. The free Bing Copilot barely gets any resources and gives very bad answers now.
If the above theory is correct (big if!), perhaps Microsoft wants to pivot to the military space. That would be in line with idealist employees leaving or being fired.
Yes, I too can see how sama could end up as Microsoft’s CEO as a result of this
For corporate use cost/benefit is a big factor, not necessarily what narrow benchmarks your expensive top model can eke out a win on.
Now I don't think that it's because OpenAI has some kind of secret sauce. It rather seems that it's mostly due to their first mover advantage and access to immense hardware resources thanks to their Microsoft partnership. Nevertheless, whatever the reason their models are superior, that superiority is quantifiable in money.
Seems more likely that OpenAI's biggest secret is that they have no secrets, and they are desperately trying to come up with a second act as tech companies with more robust product portfolios begin to catch up.
Hint: it won’t.
There are two main interpretations of what he's saying:
1) He sincerely believes that AGI is around the corner.
2) He sees that his research team is hitting a plateau of what is possible and is prepping for a very successful exit before the rest of the world notices the plateau.
Given his track record of honesty and the financial incentives involved, I know which interpretation I lean towards.
Either way he is set for life, truly being one of the most wealthy humans to have ever exist... literally.
Personally I still believe he thinks that way (in contrast to what ~99% of HN believes) and that he does care deeply about potential existential (and other) risks of ASI. I would bet money/Manifoldbux that if he thought powerful AGI/ASI were anywhere near, he'd hit the brakes and initiate a massive safety overhaul.
I don't know why the promises to the safety team weren't kept (thus triggering their mass resignations), but I don't think it's something as silly as him becoming extremely power hungry or no longer believing there were risks or thinking the risks are acceptable. Perhaps he thought it wasn't the most rational and efficient use of capital at that time given current capabilities.
Maybe he is only a greedy liar. I don't know. I'm just stating my personal belief/speculation.
Not sure how you can believe this given all of his recent actions and the ever growing list of whistleblowers dropping out of OpenAI explicitly saying Safety is not taken seriously.
I mean just generally the ability to actually stop and reorient around working on safety seems incredibly non trivial. To say nothing of the race dynamic he has perpetuated, the other frontier companies are unlikely to do the same.
That seems like fraud to me.
I can't think of a single product or company that used the "open" word for something that was actually open in any meaningful way.
So I'm assuming the game plan here is to adjust the charter of the non profit to basically say we are going to still keep doing "Open AI" (we all know what that means), but through the proceeds it gets by selling chunks of this for-profit entity, so the essence could be the non-profit parent isn't fulfilling its mission by controlling what openai does but how it puts the money to use it gets from openai.
And in this process, Sam gets a chunk (as a payment for growing the assets of the non-profit, like a salary/bonus) and the rest as well....?
Basically, the plan was to create a new for-profit entity then have the not-for-profit license the existing IP to the for-profit. There were a lot of technicalities to it, but most of that was handled by lawyers drawing up the chartering paperwork.
Safe AI, altruistic AI, human-centric AI, are all dead. There is only money-generating AI. Fuck.
so much for sam "i have no equity" altman
I'm not surprised in the least.
Who is going to give billions to a non-profit with a bizarre structure where you don't actually own a part of it but have some "claim" with a capped profit? Can you imagine bringing that to Delaware courts if there was disagreement over the terms? Investors can risk it if it's a few million, but good luck convincing institutional investors to commit billions with that structure.
At that point you might as well just go with a standard for-profit model where ownership is clear, terms are standard and enforceable in court and people don't have to keep saying "explain how it works again?".
Consider adding some EEE
It's hard to say if there is much brand value left with "OpenAI" - lots of history, but lots of toxicity too.
At the end of the day they'll do as well as they are able to differentiate and sell their increasingly commoditized products, in a competitive landscape where they've got Meta able to give it away for free.
Aaron Burr raised capital for a fake water company and applied for a banking charter for what is now JP Morgan Chase
some people are playing by a more effective set of rules and others are being lied to from a young age
An oath breaker didn’t just harm an individual, they did harm to the whole community, and the whole community viewed them negatively.
What are those books?
A slightly older but enduringly popular work is In Search of the Indo-Europeans: Language, Archaeology, and Myth (Mallory 1989). I'm told How to Kill a Dragon: Aspects of Indo-European Poetics (Watkins 1995) is very good too, but I have not read it myself.
Chapter 2 of Indo-European Language and Culture: An Introduction (Fortson 2004) is a very good summary of PIE culture. The book is excellent in general - probably the best English language introduction to PIE linguistics at the moment - but outside the first two chapters it is not generally accessible to laymen.
If you read other European languages it's worth checking out other books that may be available to you. The field by definition requires the ability to read a decent number of languages, so the literature is spread out across English, German, French, Russian, etc.
OpenAI is Microsoft's AI R&D spin-off and Microsoft means business.
The stinking peasants will never realize what's happening until it's too late to stop!
On March 1st, 2023, a warning was already sounding: OpenAI Is Now Everything It Promised Not to Be: Corporate, Closed-Source, and For-Profit (https://news.ycombinator.com/item?id=34979981)
https://www.msn.com/en-us/money/other/openai-to-become-for-p...
Text-only:
https://assets.msn.com/content/view/v2/Detail/en-in/AA1rcDWH
https://hnrankings.info/41651548/
This version of the thread is the first to have had any traction on the front page.
When the algorithm artificially stops a topic from surfacing to the front page, the article that finally makes it past the algorithm's suppression is not a duplicate, it's the canonical copy.
I'm pretty sure that what happened is that the Murati thread was id'd by the algorithm as the canonical OpenAI discussion, artificially suppressing the more interesting topic of the complete restructuring of the most important company in existence today.
I'm just glad that the Murati story falling off the front page allowed this one a second chance.
Do you have a source for this? How did they find it if not on HN?
> Merging the [dupe] only makes it better/stronger.
Moving the ~50 comments from the other thread here makes a ton of sense. All I'm saying is that this is the canonical and the other is the dupe.
Which is why we need to reopen more asylums and bring back involuntary commitment.
but still, you'd think some of them would have finally had enough and have enough opportunities elsewhere that they can leave.
"openai is nothing without its people." well, the key people left. soon, it will just be sam and his sycophants.
I get strong "next Mark Zuckerberg" vibes from Sam. Build a zombie product that approaches worthlessness after a few years, but made himself hugely rich in the process, and buys off tech and people as needed to maintain some kind of relevance.
What happened to all the people making fun of Helen Toner for attempting to fire Sama? She and Ilya were right.
Sam Altman is a poison pill.
Why the h are they called "openAI" too? nothing is open for them but your own wallet.
"Do I shock you? This is capitalism."
Yishan Wong describes a series of actions by Yishan and Sam Altman as a "con", and Sam jumps in to brag that it was "child's play for me" with a smiley face. :)
I never read that as a brag, but as a sarcastic dismissal. That’s why it started with “cool story bro” and “except I could never have predicted”. I see the tone as “this story is convoluted” not as “I’ll admit to my plan now that you can’t do anything about it”.
That’s not to say Sam isn’t a scammer. He is. It just doesn’t seem like that particular post is proof of it. But Worldcoin is.
https://www.buzzfeednews.com/article/richardnieva/worldcoin-...
https://www.technologyreview.com/2022/04/06/1048981/worldcoi...
Here is Yishan's comment with his name spelled out for clarity instead of just saying "CEO":
In 2006, reddit was sold to Conde Nast. It was soon obvious to many that the sale had been premature, the site was unmanaged and under-resourced under the old-media giant who simply didn't understand it and could never realize its full potential, so the founders and their allies in Y-Combinator (where reddit had been born) hatched an audacious plan to re-extract reddit from the clutches of the 100-year-old media conglomerate.
Together with Sam Altman, they recruited a young up-and-coming technology manager [named Yishan Wong] with social media credentials. Alexis, who was on the interview panel for the new reddit CEO, would reject all other candidates except this one. The manager was to insist as a condition of taking the job that Conde Nast would have to give up significant ownership of the company, first to employees by justifying the need for equity to be able to hire top talent, bringing in Silicon Valley insiders to help run the company. After continuing to grow the company, [Yishan Wong] would then further dilute Conde Nast's ownership by raising money from a syndicate of Silicon Valley investors led by Sam Altman, now the President of Y-Combinator itself, who in the process would take a seat on the board.
Once this was done, [Yishan Wong] and his team would manufacture a series of otherwise-improbable leadership crises, forcing the new board to scramble to find a new CEO, allowing Altman to use his position on the board to advocate for the re-introduction of the old founders, installing them on the board and as CEO, thus returning the company to their control and relegating Conde Nast to a position as minority shareholder.
JUST KIDDING. There's no way that could happen.
My understanding of what Sam meant by "I could never have predicted the part where you resigned on the spot" was that he was conveying respect for Yishan essentially out-playing Sam at the end (the two of them are friends) by distancing himself (Yishan) from the situation and any potential liability in order to leave Sam "holding the bag" of possible liability.
Looks like it was flagged, before being brought back by an admin. This happens quite often.
https://hnrankings.info/41651548/
It somehow got enough attention to get to 300+ votes while being entirely suppressed from the front page.
This one looks like it triggered a flag before being fixed:
* Frederick Banting sold the patent for insulin to the University of Toronto for just $1.
* Tim Berners-Lee decided not to patent the web, making it free for public use.
* Jonas Salk refused to patent the polio vaccine - "can you patent the sun?"
* Richard Stallman and Linus Torvalds could have easily sold humanity out for untold billions.
* Chuck Feeny silently gave away $8bn, keeping only a few million.
... And in any case, this is an extreme situation. AI is an existential threat/opportunity. Allowing it to be sidestepped into the hands of Sam "sell me your retinal scans for $50" Altman is fucking insane, and that's putting it lightly.
It's way easier to adapt an existing framework one way or the other if the political part is already done.
I don't trust the AI industry to be a good stewart even less then the tech industry in general and when the area where I live has a chance at avoiding the worst outcome (even if at a price) in this technological transition I'm taking it.
https://factmyth.com/factoids/the-robber-barons-gave-most-of...
It’s hard to find someone who has gotten to a position where they might have a reasonable shot at becoming the world’s wealthiest person who doesn’t think they’d be a great steward of the wealth. It makes much more sense for a titan of industry to make as much as they can and then give much away than it does to simply not make it.
Regardless, you've missed the point. Some people value their integrity over $trillions, and refuse to sell humanity out. Others would sell you out for $50.
Or to put it another way: Some people have enough, and some never will.
You could prove me wrong easily. Find someone who raised (inflation adjusted) tens of billions, had the opportunity to make trillions, and declined. You can’t. You can likely take that down two orders of magnitude and still not succeed.
People who did some work on their own and open sourced something small that turned into something huge are not even close to what we’re talking here. They didn’t turn down a trillion, they turned down $100k that turned into a trillion later.
It isn’t about integrity. It’s about how humans rationalize. “Nobody is being hurt here.” “This can improve humanity.” “I’ll do good things with the money.”
It’s easy to say people have “integrity” when you just define it as adhering to your belief system when the stakes seem low, not their belief system when the stakes are clearly high.
Oh, but it is. That's exactly what this is about.
As pointed out much earlier in the thread, it's generally the people who lack integrity that fail to acknowledge or recognize it in others.
"For what shall it profit a man, if he shall gain the whole world, and lose his own soul?"
- You don't have to be Jesus to understand the sense of this quote. Just be honest and observant.
They would tell you (and sincerely believe it) that it being a for profit is better for a whole list of reasons. They believe they have integrity. They don’t believe they’ve lost their soul. They believe they’re doing a whole lot of good for the world.
That’s my point. The denizens of HN think they lack integrity for this, I think they just define it differently and anyone playing for trillions would define it their way.
> And those kind of people have difficulty believing this.
No, we're really not all like that.
I stopped caring about money at 6 digits a decade ago, and I'm not even at 7 digits now because I don't care for the accumulation of stuff — if money had been my goal, I'd have gone to Silicon Valley rather than to Berlin, and even unexciting work would have put me between 7 and 8 digits by this point.
I can imagine a world in which I had made "the right choices" with bitcoin and Apple stocks — perfect play would have had me own all of it — and then I realised this would simply have made me a Person Of Interest to national intelligence agencies, not given me anything I would find more interesting than what I do with far less.
I can imagine a future AI (in my lifetime, even) and a VN replicator, which rearranges the planet Mercury into a personal O'Neill cylinder for each and every human — such structures would exceed trillions of USD per unit if built today. Cool, I'll put a full-size model of the Enterprise D inside mine, and possibly invite friends over to play Star Fleet Battles using the main bridge viewscreen. But otherwise, what's the point? I already live somewhere nice.
> There are just people who are like that and people who think they aren’t because they’ve never been within a digit grouping of it.
Does it seem that way to you because you yourself have unbounded desire, or because the most famous business people in the world seem so?
People like me don't make the largest of waves. (Well, not unless HN karma counts…)
> perfect play would have had me own all of it
I meant literally all of it: with perfect play and the benefit of hindsight, starting with the money I had in c. 2002 from summer holiday jobs and initially using it for Apple trades until bitcoin was invented, it was possible to own all the bitcoin in existence with the right set of trades.
Heck, never mind perfect play, at current rates two single trades would have made me the single richest person on the planet: buying $1k of Apple stock at the right time, then selling it all for $20k when it was 10,000 BTC for some pizzas.
(But also, the attempt would almost certainly have broken the currency as IMO there's not really that much liquidity).
Being that rich doesn't lead my imagination to new happiness that I don't already possess, only new stresses that I don't already posess. I don't want a super yacht, a personal jet, nor a skyscraper with my name on it, and a private island is less appealing to me than a tourist destination.
Knowing this about myself, I don't need to chase higher pay or timing the markets to get things I can't currently afford, instead I focus on things that I do like. Those things in aggregate cost less than €12k/year, including travel.
I don't think there's many people out there who would not be tempted at all to take some of that money for themselves. Rather, people are willing and able to rise above that temptation.
For the true believers that's just a temporary setback on the way to becoming trillionaires though.
When/if Altman ever gets out of the way like Travis K did with Uber then the real business people can come in and run the company correctly. Exactly like what happened with Uber - who never turned a profit under that leadership in the US and had their lunch eaten by a Chinese knock-off for years abroad. Can't have spoiled brats in charge, they have no experience and are wasteful and impulsive. Especially like Altman who has no engineering talent either. What is his purpose in OpenAI? He can't do anything.
Or "clopen-source software", projects that claim to be open-source but vital pieces are proprietary.
Reputationally... the net winner is Zuck. Way to go Meta (never thought I'd think this).
"OpenAI to remove non-profit control and give Sam Altman equity"
https://www.reuters.com/technology/artificial-intelligence/o...
> 57. OpenAI to Become For-Profit Company (wsj.com) 204 points by jspann 4 hours ago | flag | hide | 110 comments
I was under the impression that most people saw this coming years ago. The second "Open"AI refused to release the weights for GPT-2 for our "safety" (we can see in hind sight how obviously untrue this was, but most of us saw it then too) it was clear that they were headed towards profitability.
/s
They have that Michael Scott & Ryan energy.
I'm shocked. Shocked!
I better stock up on ways of disrupting computational machinery and communications from a distance. They'll build SkyNet if it means more value for shareholders.
And what comes first, the mission, or being able to tell people you did your best but failed to build the thing you set out to build?
Perhaps most of us are more interested in fairness than progress, and that's fine.
I'm sorry, have we gotten so far up our own asses as a profession that we no longer just excuse unethical behavior, we actually encourage it?
I feel this only scratches the surface of what to chase in life. And in respect to a potentially singular, all-knowing piece of technology, not necessarily a goal people want to embue.
If you don't mind me asking, what generation are you from? Perchance you're newer than me to Earth, among those who find it hard that others have different opinions?
Well, yeah, they're positioning themselves as some of the most powerful and influential individuals on earth. I'd say any personality flaws are pretty important.
There will always be people who disagree with the politics/opinions/alleigances of a successful person and who wish to downplay their success for selfish reasons.
And conversely, there will always be people who agree with the politics/opinions/alleigances of a successful person and who wish to overstate the reasons behind their success for selfish reasons.
If you want to understand why, read the Walter Isaacson biography of Musk (which is based on accounts by his friends, enemies and employees). He's a hard-arsed manager, he is technically involved at all levels of the company, he is relentless, and he takes risks and iterates like no other CEO.
Walter Isaacson doesn't have the best reputation for covering his subjects objectively to be fair. If your source for what Elon has or hasn't done is Isaacson, you aren't standing on very solid ground.
The bigger picture point though is that you can easily argue that the employees at those companies, and not a single man, are responsible for the success of those companies. We give far too much credit to CEOs.
Tesla began as a very early-stage startup in mid-2003 but produced nothing prior to Musk.
It was Musk that made the investment (in 2004) that Tesla needed to begin work on their first car (the Roadster). Elon Musk was co-founder, product architect and chairman of Tesla at the time of the Roadster's development. To be clear - he was the product architect of Tesla's first product, from the beginning.
Musk received the Global Green 2006 product design award for the design of the Tesla Roadster, presented by Mikhail Gorbachev, and he received the 2007 Index Design award for the design of the Tesla Roadster.
https://en.wikipedia.org/wiki/Tesla_Roadster_(first_generati...
What's being discussed in this thread is not the personal failings of silicon valley darlings, but whether one of them just defrauded a few thousand people and embezzled a significant amount of capital. Citing his character flaws goes along with it though.
Are you seriously arguing that people should be exempt from law for "advancing humanity"? Because I don't see any advancements whatsoever from all of the people mentioned. Altman and Musk would get a hardon for sure though, from being mentioned together with Jobs.
Oppenheimer "advanced humanity" by giving us nuclear power. Cool. I love cheap energy. Unfortunately, there were some uh... "unfortunate side-effects" which continue to plague us.