Checks and balances concern constraints on government power. Whether OpenAI's structure complies with the law is a question of regulation, not checks and balances.
It's not a question of regulation, but a question of enforcing said regulation and a question of enforcing a lawful process which follows a breach of a regulation. Which is exactly a spirit of the phrase "checks and balances". It means that if one branch breaks law, then the other branch enforces a compliance with laws/regulations. Of even inside the same branch, if a temporary elected doofus in charge is breaking the law, then people even lower in hierarchy will enforce normal branch functioning. Which in practice was super easy to break, sabotage and blackmail, so neither checking nor balancing happened, everyone went corrupt or impotent simultaneously.
Every step taken by the nonprofit leadership has to be, (or at least seem to be at the time), net positive for the stated goal of the nonprofit. To be legal, the IPO needs to be a net gain for the nonprofit.
It can easily be that, if they believe that the capital it raises increases the long-term value of the company by a greater multiple than the proportion of the company that is lost from the nonprofit to outside investors.
The primary example of this is Novo Nordisk (the Ozempic company). Their largest shareholder is, through an intermediary, the Novo Nordisk Foundation, which is one of the largest charities in the world. Nordisk used to be a charity that owned 100% of it's own labs and facilities, but in 1989 they realized that they were just too small, and would get trampled by larger international players without greatly increasing their scope. So they made their subsidiary go public (through a complex merger, not an IPO), and now only own 28% of it, instead of 100%. But, in large part because of the capital that going public brought them, despite constantly distributing money for research and charity, that's 28% of a company that's more than 100x bigger that what they used to be. And they retained 77% voting control.
A few things, but they work very well for our industry.
The rule is that the nonprofit and disqualified persons (mostly board members), cant own businesses together, well they can but not more than 35% of it together, and a max of 20% can have voting capability
The consequences arent immediate, non profits have 3 years to correct this
Now in the tech industry, getting VCs involved is already the plan from day one and founders get diluted, so getting below 35% is either easy, or easy within 3 years
so they’re fine
there’s a lot of things they can all do to deal with the share consolidation
If the private subsidiary was doing semi-unrelated stuff to the goals of the non-profit, and using it to fund the non-profit, then your logic could make sense - for example if a cancer research charity owned a profitable business and funnelled the profits up to spend on research, great.
But in OpenAI's case, the claimed goals of the non-profit were essentially "do AI in a way that puts safety above profits". And whether or not one agrees with their previous approach to safety, or even whether safety needs to be cared about, it's undeniable that the for-profit business isn't acting as useful fundraising for the non-profit's goals, it's literally acting in the opposite direction.
> it's undeniable that the for-profit business isn't acting as useful fundraising for the non-profit's goals, it's literally acting in the opposite direction
It's generally not up to your or to me, it's up to the donors to the non-profit. If what you find to be undeniable is very much deniable to them, then that is their right.
The only question of public concern is whether OpenAI, Inc., a charity, meets the exemption requirements [1].
1) In order to fund research - this stuff costs 10s of billions of dollars - everyone, from Ilya, to Elon, to Sam - all agreed that they would require a profit-arm to raise money. Nobody was going to sponsor that 10s of billions of dollars to a non-profit.
2) The non profit is still there - and controls the commercial element.
If they truly wanted it to be in the benefit of the not-for-profit and safe from interference, the ownership by the foundation would be much closer to or just over 50%.... just thinking out loud...
The magic 50% ownership isn't relevant for that purpose. There are special provisions which means that the Foundation effectively exerts full control over the company because it appoints the entire board.
The corporation selling shares is just primarily owned by the non profit
The corporation selling shares is subject to normal corporate tax regime
The real answer to your question is that non profits can own shares, and there is no legal difference between passive investment of other publicly traded companies and highly consolidated shares of a private company. In the US it is seen as merely happenstance that we have such a liquid market where the shares themselves can rapidly change in value and create profits, but there is nothing controversial about that.
Why not finance a more wide tax break for tech companies, including small companies?
I have seen first hand, one of my classmates from high-school, whom I didn't consider to be too bright, receive a 100k EUR grant to build an esports platform that ended up being a styled-up wordpress blog. He had zero interest in tech, never programmed, the works...
He had no interest nor knowledge relating to tech, but managed to somehow get that grant. Meanwhile, I was paying taxes on hard earned freelancing dev money. We were both 23 years old at the time and it was really jarring.
I'd rather they cut taxes for already profitable small companies. The taxes in EU are astronomic.
Tax breaks is not going to do much, because that is not where the real problem is. Imo EU is being underestimated. The EU is really competitive at boring high tech that bootstrap themselves like cars, civilian air planes and tooling, so the business and innovation climate is not bad.
What's different from the EU and US is that the US has single capital market with more unified regulations where it is much easier to pool infinite VC money into an idea. Contrary to popular belief it's actually the LACK of EU regulations that makes the EU less competitive, because a company has to deal with 20+ regulatory bodies.
In short, the EU should focus on more market integration.
Yep, so much this (and Draghi concurs). It's kind of bizarre that the EU needs to do what it's biggest critics fight the most: integrate it's markets and eliminate cross-member bureaucracy.
> that bootstrap themselves like cars, civilian air planes
A civilian airplane manufacturing business takes decades to get off the ground, and is practically impossible to start without heavy government involvement. Cars are not that far behind. How did you pick these as businesses that "bootstrap themselves"? Bootstrapping requires a non-capital intensive business model. These are the opposite. These are businesses running on decades of momentum and government support.
Probably should say companies that are not financed with infinite VC company. Since that is a uniquely American advantage that is not developed in Europe
yet.
You mean the cars whose manufacturing has to be stopped because they don't know how to use one chip in pace of another, while Tesla was chugging in numbers because they were the only ones who knows how to program a new chip.
Tesla is one of the few car companies that receive infinite VC money. European cars are more competitive then American. There is a reason why Trump is complaining that no country is buying in American cars. There are far more European cars in Asia than American ones.
Because investment is often tax-deductible anyway now-days with modern policy trying to encourage this via huge tax-discounts on capex. Therefore tax-breaks on profits just allow existing owners to cash out cheap. The companies that need the help need capital, not tax discounts. Further to this the companies that are profit-generating, still require capital for growth, the small tax on profits isn't enough to make a difference in the scale of things for investment required since corporation tax rates are often so low.
Can the EU directly regulate the details of the tax regulations of their member states? That would be extremely surprising to me and, even if done, would likely be seen as a sever overreach.
It's quite simple. The EU can't tell member states how much to tax and even if you cut out all of its budget you'd save maybe a percentage point or two in taxes
On the other hand the EIB (which is anyway not directly controlled in their executive actions by the EU) can grant funds since it's kind of it's role as an investment bank
In Norway, you can get 25% tax break on R&D costs. It can be combined with other softfunding up to 70% for small companies doing research, and 50% for larger companies doing development.
but how does that work? The money you pay on taxes, is on income - that money is now yours. On the other hand I'd assume a grant is the property of the company, not your friend. Where they able to just leach it all away as personal income?
A rare game where having worse PC actually made you perform better. It was easier to time the jump perfectly. I recall starting 5 instances of it to slow it down as a kid and beating records. Fun times.
The United States debatably has the most athletic population of an country and its the top sport in America where we funnel all of our talent (probably to our detriment). Regardless of the popularity abroad its where our athletes go.
When we do compete in other sports we fair above average to exceptional (https://en.wikipedia.org/wiki/All-time_Olympic_Games_medal_t...) with the notable exception of soccer, which we've never broken through despite investment and a decent population of players.
Don't kid yourself, just because other countries don't play American Football doesn't mean the players aren't freak athletes - they are
It’s not the #1 sport there, but NFL football is popular in Canada too. To me, it’s dominance in North America is enough to qualify but I respect your position.
So some CPU models wait 120 cycles and others wait a millisecond (millions of cycles)? That seems like a pretty drastic difference. I wonder why they would write code with such a drastic difference between CPU models.
The nice thing about compiler optimizations is that you can improve performance of existing CPU's without physically touching them. Year by year. You squeeze more of the machine someone designed. It adds up.
Imagine what environmental impact you would have if you optimized Python's performance by 1%? How much CO2 are you removing from the atmosphere? It's likely to overshadow the environmental footprint of you, your family and all your friends combined. Hell, maybe it's the entire city you live in. All because someone spent time implementing a few bitwise tricks.
> Imagine what environmental impact you would have if you optimized Python's performance by 1%?
I imagine this would also increase the complexity of the python intepreter by more than 1%, which in turn would increase the number of bugs in the interpreter by more than 1%. Which would burn both manpower and CPU-Cycles on a global scale.
(I assume that optimization, that reduce complexity are already exhausted, e.g. stuff like "removing redundant function calls". )
This is an assumption that a reasonable person naively comes up with.
Then if you actually go ahead and check, it turns out it's not true! It's quite a shocking revelation.
When you dig into the popular compilers/runtimes (with the exception of things like LLVM)
Many of them still have low hanging fruit of the form:
a = b + c - b
Yes, the above is still not fully optimized in the official implementations of some popular programming languages.
Also an optimization of "removing redundant function calls" isn't a binary on/off switch. You can do it better or worse. Sometimes you can remove them, sometimes not. If you improve your analysis, you can do more of that and improve performance. Same for DSE, CSE, etc...
In many languages, you can't just optimize + b - b willy-nilly, as there could be side effects and non-obvious interactions abound. For instance, in JavaScript, where everything is a 64-bit double, a + b - b is definitely not the same as a, given large enough or small enough a or b. In LLVM for floats as well, certainly.
It's an example of how trivial some of this low hanging fruit is, the above is a concrete case that I have personally implemented (arithmetic of in-register 64/32-bit integers). You can get into semantics and restrictions, but I think the point I'm raising is clear.
reply