Also relevant: The Trouble With Fusion [1]. It's from 1983, but could well have been written today, as nothing has really changed in the grand scheme of things. In short, there are multiple technical and economical challenges that stand in the way of nuclear fusion becoming a viable path towards clean energy, even after the first successful demonstration of a self-sustained burning plasma. Which, by the way, still seems about 10 years out, with the latest round of delays on ITER. I think most of the people involved understand this, but there is ongoing political will to sink billions of dollars into these projects, so here we are.
Lidsky is long dead (as are Pfirsch and Schmitter, who had similar critiques in Germany), but I like to think he'd have been a supporter of Helion. Helion's approach is close to what Lidsky advocated (aneutronic fusion), at least before Lidsky's student Todd Rider showed truly aneutronic fusion was not likely to work. However, the Helion approach has greatly reduced neutronicity, particularly of energetic DT neutrons, and exploits the kind of non-Maxwellian plasmas w. energy recovery/recirculation that Rider was analyzing.
There actually is still significant lift. We define the edge of the atmosphere to be where the lift to drag ratio of a plane would be less than 1 below orbital velocity (ie if you were going fast enough to lift your weight with conventional wings you'd be in orbit), so you can't fly conventionally in space but lift might still be generating a force which is significant compared to your craft's weight.
Well the assumption was that there is no drag because the air density is so low. You can’t just say there’s no drag but still assume that you get lift. Your lift/drag ratio won’t go up infinitely just because you’re flying higher.
But judging by "in four hours" I'm guessing he's imagining something somewhere in between those two extremes. High enough to substantially reduce drag, low enough that you don't need to approach orbital velocity to maintain altitude.
"Fast enough" is very nearly orbital speed, though. Suborbital range is very short on the lower end, and increases rapidly and nonlinearly later. E.g. if you can boost to 2km/s (~ Mach 7), this gives you, I kid you not, around 200km of ballistic range. It's either atmospheric flight or orbital flight, and there's nothing really useful in between.
The minimum amount of work needed to pump some amount of heat Q from a temperature T0 to a higher temperature T1 is W = Q*(T1/T0 - 1). For example, if your ambient heat sink is at 20C (293K) you need at least 2.8W of electricity to run the cooler for every 1W dissipated at 77K, or 28.3W for 1W dissipated at 10K. This is the thermodynamic lower limit, and practical heat pumps will be less efficient in general. In practice it might be something like 4x and 50x, respectively.
Apart from my general queasiness about the whole AGI scaling business and the power concentration that comes with it, these are the exact four people/entities that I would not want to be at the tip of said power concentration.
<quote>
Citizens will be on their best behavior because we are constantly recording and reporting everything that's going on," Ellison said, describing what he sees as the benefits from automated oversight from AI and automated alerts for when crime takes place. "We're going to have supervision," he continued. "Every police officer is going to be supervised at all times, and if there's a problem, AI will report the problem and report it to the appropriate person.
</quote>
What is far more important to understand is to ignore all that nonsense and focus on who makes money? It will be Ellison and his buddies making tens of billions of dollars/year selling 'solutions' to local governments, all paid by your property taxes. This also enables an ecosystem of theft, where others benefit a lot more. With the nexus of Private Prisons, kids for cash judges (or judges investing in stock of prisons), DEA/police unions, DEA unions, small rural towns increasing prison population (because they get added to the total pop, and get funds allocated).
More importantly this is extremely attractive to police who can steal billions every day from civil forfeiture, they have access to anyone who makes a bank withdrawal or transacts in cash, all displayed in real time feeds, ready for grabbing!
> "Citizens will be on their best behavior because we are constantly recording and reporting everything that's going on," Ellison said, describing what he sees as the benefits from automated oversight from AI and automated alerts for when crime takes place.
Wow! It is genuinely frightening that these people should be in control of our future!
Literal 'new world order' stuff here. Alex Jones and crew got so excited that their guy was in the driver's seat that they didn't notice the actual illuminati lizard people space lasers being deployed.
I don't think we'll ever have a zero-crime society, neither should we aim to be one. But being left to the vagaries of police (and union) politics, culture and the complications of city budgets is clearly broken.
Example: Cities are being presented a false choice between accepting deadly high speed chases vs zero criminal accountability [1], which in the world of drones seems silly [2]
I don't want the police to have unfettered access to surveil any and all citizens but putting camera access behind a court warrant issued by a civilian elected judge doesn't feel that dystopian to me.
Is that what Ellison was alluding to? I have no idea, but we are no longer in a world where we should disregard this prima facie.
If you look at the original video [1], starting at 1:09:00, he's talking specifically about police body/dashcams recording interactions with citizens during callouts and stops, not everyone all the time as that article strongly implies. The USA already decided to record what police see all the time during these events, so there's no new privacy issue posed by anything he's suggesting. The question is only how those videos are used. In particular, he points out that police are allowed to turn off bodycams for privacy reasons (e.g. bathroom breaks), which is a legitimate need but it can also be abused, and AI can fix this loophole.
In the same segment he also proposes using AI to watch CCTV at schools in real time to trigger instant response if someone pulls out a gun, and using AI to spot wildfires using drones. For some reason the media didn't condemn those ideas, just the part about supervising cop stops. How curious.
Move to China. You’ll love it. Not only does it have lower crime by virtue of being highly controlled, it also has the added benefit of you never hearing about crimes the government doesn’t want you to hear about, and you won’t hear about any police corruption or brutality either. Ignorance is bliss!
You also don't need to sell your company when you innovate, your well-connected oligarch has either already gotten what you're doing or can seize it without consequence.
Philosophically I agree, this sounds nice. A bit false dichotomy-ish, but nice.
But if you think about it, an unstated yet necessary prerequisite is that the definition of "crime" must be morally aligned with what is right. If it's not, well then you're living in a dystopia. Imagine a world where slavery is still legal and being a runaway slave is a crime. How do people like Frederick Douglass escape and survive long enough to make a difference?
And that's before we get into the prerequisite that such a state must apply the laws completely evenly with no special tiers based on class, wealth, political connection, celebrity status, etc, which AFAIK has never been done. Given the leadership, it doesn't look like it's goig to happen anytime soon. IMHO I think it's heavily contrary to human nature and just won't be achievable short of altering human nature.
Even if shoplifting at Walgreens was the reason for closure, the downtowns of a few "liberal cities" (it's always the same 3-4 mentioned) are extremely unlikely to have that much impact on national statistics.
The retail industry lobbying group itself noted that "shrink", the term for loss of revenue due to items walking off or being damaged, has remained unchanged since the 90s.
The people telling you that there is an immense wave of shoplifting are outright lying.
There's a few that have tried to implement this, and I want to live in none of them.
The US will fare no better if it walks down this path, and honestly will likely fare worse for it's cultural obsession with individualism over community.
If you're lucky, you might get your chance to live in Thiel's and Ellison's techbro utopia. Make sure to tell us how great it is to be subjected to people with no accountability, but all of the power over every aspect of your life.
Just Ellison alone brings unwelcome feeling of having Oracle craziness forced down our collective throats, but I share your concern about the unholy alliance generated in front of us.
My immediate reaction to the announcement was one of these is not like the others. OpenAI, a couple of big investment funds, Microsoft, Nvidia, and...............Oracle?
Oracle provides two things: A datacenter for Nvidia chips, and health data. Oracle Cerner had a 21.7% market share for inpatient hospital Electronic Health Records (EHR). Larry Ellison specifically mentioned healthcare when announcing it in the Whitehouse.
The announcement was funny because they weren't quite sure what they are going to do in the health space. Sam Altman was asked, and he immediately deferred to Ellison and Masayoshi. Ellison was vague... it seems they know they want to do something with Ellison's massive stash of health data... but they don't quite know what they are building yet.
The Snowflake-for-health is more about opening EHR data for operational use by providers and facilities.
Versus being locked into respective EHR platforms.
If Oracle provided a compelling data suite (a la MS) within their own cloud ecosystem, they'd have less reason to restrict it at the EHR level (as they'd have lock-in at the platform level), which would help them compete against Epic (who can't pivot to openness in the same way, without risking their primary product).
I think you mean PostgreSQL for EHR data. MS Fabric and Snowflake are analytical databases, not operational. Patient privacy requirements (and HIPAA law) is a blocker for having an open operational database for EHR.
Oracle makes perfect sense in that they are 1) a massive datacenter company, and 2) sell a variety of saas products to enterprises, which is a major target market for AI.
> Oracle has 2-3% market share as a Cloud Provider.
And the market leader is what, 30%? about 1 order of magnitude. That's not such a huge difference, and I suspect that Oracle's size is disproportionate in the enterprise space (which is where a lot of AI services are targeted) whereas AWS has a _ton_ of non-enterprise things hosted.
In any case, 2-3% is big enough where this kind of investment is 1) financially possible, 2) desirable to grow to be #2 or #3
Getting from 2% (Oracle) to 10% (GCP) market share would need 37.97% CAGR in 5 years. In a vacuum where everything else keeps the same, maybe, but I see that goal as very difficult to attain in what is a highly competitive industry right now.
Disclaimer: I work at a highly regulated industry and we are fine running our "enterprise" workloads in Azure (and even AWS for a spinoff company in the same sector). Oracle has no specific moat in that area imho, unless you already locked-in in one of their software offerings.
There is a certain reason that last weeks everybody and their grandma is simping for Trump. Nobody would want to be on his bad side right now. Moreover, we hear here and there that Trump "keeps his promises". A lot of the promises we do not know about and we may never will. These people did not spend money supporting his campaign for nothing. In other places and eras this would have been called corruption, now it is called "keeping his promises".
Trump is one of the most famous people in the world for not keeping promises of paying debts. But there is money to be made temporarily when he is running a caper, as long as you can get your hand in the pot before he steals it.
If your knee jerk response to any political discussion even remotely critical of 'your guy' is to snap into whataboutisim instead of participating in the conversation you might need a outrage pornography detox for a while.
> There is a certain reason that last weeks everybody and their grandma is simping for Trump. Nobody would want to be on his bad side
It's worth keeping in mind how extremely unfriendly to tech the last admin was. At this point, it's basically proven in court that emails of the form "please deboost person x or else" were send, and there's probably plenty more we don't know about.
Combine that with the troubles in Europe which Biden's administration was extremely unwilling to help with, the obstacles thrown in the way of major energy buildouts, which are needed for AI... one would have to be stupid to be a tech CEO and not simp for Trump.
Tech has been extremely Democratic for many years. The Democrats have utterly alienated tech, and now they reap the consequences.
> Tech has been extremely Democratic for many years. The Democrats have utterly alienated tech, and now they reap the consequences.
Well, on the other side it can be said that Big Tech wasn't really on the side of democracy (note: democracy, not the Democrat Party) itself, and it hasn't been for years - at the very least ever since Cambridge Analytica was discovered. The "big tech" sector has only looked at profit margins, clicks, eyeballs and other KPIs while completely neglecting its own responsibility towards its host, and it got treated as the danger it posed by the Biden administration and Europe alike.
As for the cryptocoin world that has also been campaigning for the 45th: they are an even worse cancer on the world. Nothing but a gigantic waste of resources (remember the prices of GPUs, HDDs and RAM going through the roof, coal power plants being reactivated?), rug pulls and other scams.
The current shift towards the far-right is just the final masks falling off. Tech has rather (openly) supported the 45th than to learn from the chaos it has brought upon the world and make at least a paper effort to be held accountable.
Yes, big tech was the kid caught in the corner cleaning out the cookie jar and threw a tantrum when one parent moved the jar out of reach as punishment in effort to help the industry learn self-control. Now the other parent has come home and has not only returned the cookie jar to the kid but pledged to bring them packs of cookies by the shipping container to gorge on in exchange for favors.
Nice euphemism for giving people autonomy in their data and privacy.
Most of there companies are so large that they cannot really fail anymore. At this point it has very little to do with protecting themselves, more with making them more powerful than governments. JD Vance are said that the US could drop support for NATO if Europe tries to regulate X [1]. Oligarchs have fully infiltrated the US government and are trying to do the same to other countries.
I disagree with the grandparent. They don't support Trump because they do not want to be on his bad side (well, at least not only that), they support Trump because they see the opportunity to suppress regulation worldwide and become more powerful than governments.
We just keep making excuses (fiduciary duties, he just doesn't know how to wave his arm because he's an autist [2]). Why not just call it what it is?
I do agree that big part of why they support Trump is for anti-regulation reasons. But, it is also a fact that Trump is one of them, a businessman, not a politician. With Trump they can now discuss more business and less policies. There is a certain dealing of business right now that seems not at all transparent. And in this, the amount of public simping is really weird to what usually happens, everybody praising Trump even before he was taking office, and even tiktok, "coming out" as whatever etc.
Oligarchs want less regulation, but they also want these beefy government contracts. They want weaker government to regulate them and stronger government to protect them and bully other countries. Way I see it, what they actually want is control of the government, and with Trump they have it (more than before).
We have more energy and are pumping more domestic oil than ever. We are a major exporter of LNG. Trump just killed EV subsidies, and electric charging network funding.
What are you talking about via Europe? Holding tech companies accountable to meddling in domestic politics? Not allowing carte blanche to user data?
I understand (though do not like) large corps tiptoeing around Trump in order to manipulate him, it is due to fear. Not due to Trump having respectable values.
Tax breaks, government forced to become a customer etc. the usual. Just like the astronauts to Mars thing will just shovel your money that might have gone to NASA into Musk's pocket.
> the usual. Just like the astronauts to Mars thing will just shovel your money that might have gone to NASA into Musk's pocket.
The difference is that Musk can do twice as much for 1/10 what Nasa thinks the program will cost, which is never what the program will actually cost, and Musk will do it in half that time to boot.
The guy is an unhinged manchild, but if what you care about is having your money well spend and getting to Mars as cheaply as possible, he's exactly who you're looking for.
I'm sure you are claiming that the founder, CEO, and controlling shareholder is "kept a careful length from the controls" because you have detailed first hand knowledge of the internal operations of Space, right?
> if what you care about is having your money well spend and getting to Mars as cheaply as possible, he's exactly who you're looking for.
I do find impressive that SpaceX engineers figured out reusable rockets and now we can send things more cheaply out to orbit. But in all seriousness, should we care about getting to Mars cheaply? Or do people care because Musk came along to convince them (and the US government) to invest in this venture of his?
Tax breaks, i.e. a company extracting wealth from a community without paying into the systems that keep all the parts of that community running, forcing the community to ultimate subsidize that business's weath extraction from them.
Companies do not extract value, they create value which is then transferred to the people via the market through voluntary exchange (ideally). Where have you learned about those things? Oh, yeah, “community” , i.e. Marx.
Yes, before the resource is taken out of the Earth it doesn’t exist, it is created in a sense by them. Look at Venezuela - they are dying of hunger with all the oil in the world (Russia, too) But socialist ideas prevailed there and the bad companies are banned.
Assuming the tax money has to come from somewhere at some point, those who pay taxes have to make up the shortfall from those who have tax breaks. So far the US just kicks that can down the road so...
That is a big assumption. Tax money need not be a constant. But for the sake of following the same logic: if companies pay bigger taxes, they also have to make up the shortfall. Actually, this last one is much more accurate statement. Companies do not pay taxes, PEOPLE pay taxes. So taxes are paid either by the employees, the clients or by the owners (which in case of the big tech are generally common people). With high taxation you are hurting: the customers, the workers and the middle class saving for their retirement. Who is winning the tax money: state bureaucracy, corrupt politicians and the business around them, people who live like parasites (or rather are forced to live like that, because they are electoral power).
This is utter nonsense. If 1000 people go to a deserted island with no government and taxation would that mean the inflation will be plus infinity or at least very high??? Inflation is monetary phenomenon, it happens when money is being printed.
In that case there would be no inflation or deflation, assuming a fixed money supply and no economic growth. However, the the key here is that the government, the federal government anyways, is spending money regardless of the tax break. Anytime the government writes a check, that's a little bit more money floating around; anytime the government collects some money, such as taxes, there's that much less money to be had. Every tax break causes the money supply to increase more relative to if the tax break did not exist, causing more inflation (or less deflation, if that were the case). If the government spent exactly as much as it taxed, then there would be... actually deflation, because the economy is growing. This is the basics of fiscal policy.
There's also the monetary policy, which is when the federal reserve does this on purpose. The general principle is the same, but instead it spends its money buying bonds and gets its money selling those bonds, and creates a bunch of rules about where banks keep their money so it always has some money on hand.
So, in this desert there would be no inflation or deflation, you say. Let’s say we use gold coins there. Wouldn’t we have an inflation if we find a gold mine there and everybody start digging up gold?
You are missing the fact that the money printing is not driven only by government spending. It is driven primarily by the monetary policy (in the hands of the FED) and to some extent by the government debt. You have knowledge gaps on a very basic level.
The idea that taxation stops inflation is absolutely ridiculous. It would mean that countries with low taxes have very high inflation and this is not the case. It would also means that the inflation should be constant and in struct correlation with the taxes. Both statements are completely false and very easily provable by quick fact check.
The only things taxes do are: misplacing capital and stopping economic growth, which may be the same thing arguably
> It would mean that countries with low taxes have very high inflation and this is not the case.
It's about the total balance of government spending and taxes. The point being made is that tax breaks have the same effect as government spending. Recall that I was replying to
> Tax breaks, i.e. my money not being in your pocket means that they are stolen?
The government writing someone a million dollar check and the government giving someone a million dollar tax break (assuming they pay at least a million in taxes), contribute to inflation by increasing the money supply by a million dollars than it would be otherwise. Yes, this federal reserve is by far a larger driver of inflation, but the government giving this tax break still degrades the value of your money, same as if they wrote a check.
Of course, it is easy to view a tax break as a non-action, but that's exactly why the government gives so many tax breaks. Once you're taxing everyone, you can hand out tax breaks that's the same as handing out money only you can pretend that it's doing nothing.
Think of it as 3 Scenarios:
1) The island government writes a check to everyone except you, increasing their wealth by 50%.
2) The island government taxes just you for 50% of your wealth.
3) The island government taxes everyone 75% of their wealth, grants everyone but you a total tax-break, and you 25 percentage point tax break.
Basically the same result, only in one they say "It was fair, and we handed out a few tax-breaks, what's wrong with letting people keep their money?"
1) wealth is not increased by tax break, only current income is increased
2) if government gives everyone tax break but not me, it means only that the government taxes only me
3) if everyone has 50% more money, there is very high probability that my business will go up A LOT
Seriously, dude, it’s not worthy anymore to try and explain to you very basic stuff. Inflation is not a balance between taxation and spending. All Middle Eastern countries are having huge spending and almost zero taxes. I asked you very simple question and you couldn’t answer.
What bothers me most is why people write about things they have no clue about and clearly haven’t even put a decent thought into it.
Basically what you believe in is that the thieves are controlling the inflation because they get some of the citizens wealth.
There's actually lots of taxes that aren't income or sales tax
> if everyone has 50% more money, there is very high probability that my business will go up A LOT
No, you'll be getting twice the money, but the money is worth half as much.
> Inflation is not a balance between taxation and spending.
It is for the US federal government.
> All Middle Eastern countries are having huge spending and almost zero taxes.
Those countries peg their currency to the dollar. Their money doesn't come from taxes, but instead from state oil companies. These countries aren't as free to hand out money like the US. If enough people tried to exchange their Saudi Riyals for dollars quick enough, and the Saudi government couldn't gather US dollars quick enough, their currency would very quickly collapse.
Mostly benefiting the fossil fuel industry. How are they going to power this? Gas is the only option that can be implemented within single years. And this is going to need a lot of power.
There probably will be a clause of mandatory consumption of a given percentage of power generated from coal ensuring continued coal generation of a given minimum providing excellent talking-points for broadcasting to the incumbent's base.
Maybe at some point they are going to AI themselves out of climate change. Well.. except for the part where they don’t believe in man-escalated climate change.
> Offshore wind is among the sources of new power generation that will cost the most, at about $100 per megawatt hour for new projects connecting to the grid in 2028, according to estimates from the Energy Information Administration. That includes tax credits under the Inflation Reduction Act, which reduces the cost of renewable technologies. But onshore wind is one of the cheapest sources, at about $31 on average for new projects.
I mean, I can see how numbers wise this decision makes sense.
You need to stop this nonsense. Pollution is a long term problem, but it does not mean it is productive to do what Germany has done and cease development.
It won't unless there's another (r)evolution in the underlying technology / science / algorithms, at this point scaling up just means they use bigger datasets or more iterations, but it's more finetuning and improving the existing output then coming up with a next generation / superintelligence.
There is lots of human feedback. This isn’t a game with an end state that it can easily play against itself. It needs problems with known solutions, or realistic simulations. This is why people wonder if our own universe is a simulation for training an asi.
> Being pessimistic, how come no human supergeniuses ever took over the world? Why didn't Leibniz make everyone else into his slaves?
We already did. Look at the state of animals today vs <1 mya. Bovines grown in unprecedented mass numbers to live short lives before slaughter. Wolves bred into an all new animal, friendly and helpful to the dominate species. Previously apex predators with claws, teeth, speed and strength, rendered extinct.
Sometimes I wonder if we are going to be the unkillable plague that takes over the universe. Or maybe we will dissappear in a blink. It's hard to know, we don't have any reference point except ourselves.
Once you have one AGI, you can scale it to many AGI as long as you have the necessary compute. An AGI never needs to take breaks, can work non-stop on a problem, has access to all of the world's information simultaneously, and can interact with any system it's connected to.
To put it simply, it could outcompete humanity on every metric that matters, especially given recent advancements in robotics.
...so it can think really hard all the time and come up with lots of great, devious evil ideas?
Again, I wonder why no group of smart people with brilliant ideas has unilaterally imposed those ideas on the rest of humanity through sheer force of genius.
An equivalent advance in autonomous robotics would solve the force projection issue, if that's what you're getting at.
I don't know if this will happen with any certainty, but the general idea of commoditising intelligence very much has the ability to tip the world order: every problem that can be tackled by throwing brainpower at it will be, and those advances will compound.
Also, the question you're posing did happen: it was called the Manhattan Project.
So don't plug the smart evil computer into the strong robots? Great, AI apocalypse averted.
The Manhattan Project would be a cute example if the Los Alamos scientists had gone rogue and declared themselves emperors of mankind, but no, in fact the people in charge remained the people in charge - mostly not supergeniuses.
Quite a few have succeeded in conquering large fractions of the Earth's population: Napoleon, Hitler, Genghis Khan, the Roman emperors, Alexander the Great, Mao Zedong. America and Britain as systems did so for long periods of time.
All of these entities would have been enormously more powerful with access to an AGI's immortality, sleeplessness, and ability to clone itself.
Alexander the Great made his conquests by building a really good reputation for war, then leveraging it to get tribute agreements while leaving the local governments intact. This is a good way to do it when communication lines are slow and unreliable, because the emperor just needs to check tribute once a year to enforce the agreements, but it's weak control.
If Alexander could have left perfectly aligned copies of himself in every city he passed, he could have gotten much more control and authority, and still avoided a fight by agreeing to maintain the local power structure with himself as the new head of state.
Oh, you're assuming an entire networking infrastructure as well. That makes way more sense, but the miracle there isn't AGI - without networking they'd lose alignment over time. Honestly, I feel like it would devolve in a patchwork of different kingdoms run by an Alexander figurehead... where have I seen this before?
The problem you're proposing could be solved via a high quality cellular network.
1)Leibniz wasn't superhuman
2) Leibniz couldn't work 24/7
3) he could not self increase the speed of his own hardware (body)
4) he could not spawn 1 trillion copies of him to work 24/7
Like how much time did you think before writing this
A trillion hyperintelligent demons might be cogitating right now on the head of a pin. You can't prove they aren't thinking up all sorts of genius evil schemes. My point is that "intelligence" has never been a sufficient - or even necessary - component of imposing ones will on humans.
I feel like HN/EA/"Grey Tribe" people fail to see this because they so worship intellect. I'm much more likely to fall victim to a big dumb man than smart computers.
huh what ? this is a whole new level of flat earther thinking. you actually believe that apex predators are not the most intelligent? Like humans became the apex because of something else than intelligence? AI and covid showed humanity what levels of wacky stuff people believe. I am not trying to convince you, thank you for showing me this perspective :)
I consider many successful military leaders and politicians to be geniuses as well. In my books, Caesar is as genius as Newton!
Having said that, we do not to understand the world to exploit it for ourselves. And what better way to understand and exploit the universe than science? Its an endearment.
"this generation shall not pass"... to me that's about as credible as wanting to "preserve human consciousness" by going to Mars.
Setting the world on fire and disrupting societies gleefully, while basically building bunkers (figuratively more than literally) and consolidating surveillance and propaganda to ride out the cataclysm, that's what I'm seeing.
And the stories to sell people on continuing to put up with that are not even good IMO. Just because the people who use the story to consolidate wealth and control are excited about that, we're somehow expected to be excited about the promise of a pair of socks made from barbed wire they gave us for Christmas. It's the narcissistic experience: "this is shit. this benefits you, not me. this hurts me."
One thing is sure, actual intelligence, regardless of how you may define it, something that is able to reason and speak freely, is NOT what people who fire engineers for correcting them want. It's not about a sort of oracle for humanity to enjoy and benefit from, that just speaks "truth".
The military project is/was JWCC, TS/SCI+ classified, air gapped clouds. Google, Amazon, Microsoft, Oracle etc all have existing contracts with the DoD covering such work (and getting the contract meant building and running a cloud, not just signing some paperwork agreeing to at some point in the future)
The data centres were already being built. All of these companies have been dumping tonnes of money into AI and will continue to dump tonnes of money into AI. It's just more of the same, but they had to do a big announcement with Trump to pander to his ego and somehow make it about him. Like he engineered this Stargate thing. The whole embarrassing spectacle was likely arranged by Ellison.
I was bullish on OpenAI, but honestly I don't see any path forward where they have any differentiating value that justifies even a tenth of the valuation. Their video AI is simply terrible. Dall-E 2 is matched by many competitors. 4o and o1 and good, but already have been eclipsed by a number of competitors, including an open source Chinese option.
My work has almost entirely transitioned to competitors, and Google's latest updates have quietly absolutely trounced OpenAI's offering. Like, Gemini has quietly become the best AI platform in the game.
That's all neither here nor there, but I just don't care what Altman and crew have to say any more. They are not leaders in the space. They are, in many ways, has beens.
of course. its an arms race by definition so its all a military project. and already one whistleblower was brazenly murdered by our government to protect our horse in this race.
Regarding to your question, yes. I'd prefer a healthy counterbalance to what we have currently. Ideally, I'd prefer cooperation. A worldwide cooperation.
Arguably the cooperation between the US and China has lead to the most economic growth and prosperity in human history, it's a shame the US and China are returning to a former time.
From what I've read about DeepSeek and its founder, I would very much prefer them, even with China factored in. At least if these particular Four Horsemen are the only alternative.
On a tangential note, those who wish to frame this as the start of the great AI war with China (in which they regrettably may be right), should seriously consider the possibility of coming out on the losing end. China has tremendous industrial momentum, and is not nearly as incapable of leading-edge innovation as some Americans seem to think.
No, I was rather pointing out that getting into an altercation that you are likely (even if not guaranteed) to lose may not be the smartest of ideas. On occasion, humans have been known to fruitfully engage in cooperation and de-escalation. Please pardon my naive optimism.
"Great AI war with China", "altercation" are excessively harsh characterizations. There is nothing "escalatory" in competing for leadership in new industries with other states, nor should it be "regrettable". No one, to my knowledge, is planning to nuke DeepSeek data centers or something.
I wish I could agree with you. But have you read Aschenbrenner's "Situational Awareness" [1]? I am very much afraid that the big decision makers in AI do in fact think in those terms, and do not in any way frame this as fair competition for the benefit of all.
A person heavily invested in this wave of AI succeeding saying AI will be big and we will have AGI next year? Sure.
I don't think there is much point of reading the whole thing after the following:
"Everyone is now talking about AI, but few have the faintest glimmer of what is about to hit them. Nvidia analysts still think 2024 might be close to the peak. Mainstream pundits are stuck on the willful blindness of “it’s just predicting the next word”."
China is much more peaceful nation compared to US. So, yes, I'd prefer China leading AI research any day. They are interested in mutual trade and prosperity, they respect local laws and culture, all unlike US.
I encountered this almost first person. When American company goes like an elephant, bribing local officials left and right, using dirty practices to push out concurrents. At the same time, Chinese companies try very hard to abide to local regulations and trying to resolve all issues using local courts, etc. Like actually civilised people.
What happens inside China is nothing of my interest, it's their business. They existed for millennias, they probably know how to manage themselves. They are not trying to expand outside of may be Taiwan, they don't put their military bases in my country, they don't fund so-called "opposition" and that's good enough for me.
Bribery is probably one of the few cases where the US is significantly better than bad actors in both China and the EU, both of which have major problems with overseas bribery
If you had AlQaeda in a hypothetical region near Florida with almost two-yearly terror attacks, you would shit bricks and
create jails/prisons with more security than the Pentagon itself.
I think there's a more nuanced version of this: China respects local laws and culture _outside of what they view as China_ more than the US does. It's also worth noting that China's policy in Xinjiang is somewhat narrowly targeted at religion, and less other aspects like cuisine or clothing. That said, religion is nigh impossible to separate from the broader idea of culture in much of the world.
Give me a break. China has overseas police stations as bases of operation for harassing ex-pats and dissidents. That's not "respecting local laws and culture".
Agree, and would like to say that this is not because many of us see China as some benevolent actor on the world stage, but rather because we see how NOT benevolent the US has been historically (see South America, the middle east, etc)
the US has done lots of positive things as well. i understand it is popular to be a critic nowadays, but in many ways the US has had a strong commitment to majoritarian democracy over the last century and is trending in a better direction.
but regardless of the net balance of actions, it is clearly more interventionist than China has been up to this point
Holy smokes. Do folks like you actually believe this? China has its own style of colonialism (whatever you want to call it) but it certainly exists as strong as the US flavor.
Quite a few from an economic perspective. Like I said they have their own style of colonialism. To think they are some peaceful loving nation is foolish. Maybe in the last 10 years China have had the military equipment capable of handling an offensive. They have been smart and done all their dealings via money. Without going too far in whataboutism, I simply find it ridiculous to classify China as a warm fuzzy nation with their long list of human rights issues. That does not mean America is peaceful and loving, simply that perhaps the two countries are not so different in net.
They were asking about bombs and invasions in the literal sense, not metaphorical. I'm sure if you asked someone in Gaza or Iraq if they would rather be subjected to more of America's bombing and war crimes, or China's abstract, metaphorical "economic colonialism" they would pick China in a heartbeat.
if Gaza was neighboring China and treated China like it does its neighbors, it would be treated quite differently by China. For an illustrative example, see Uighurs.
Uyghurs committed frequent acts of terrorism against China throughout the 1990s and 2000s, so you actually don't need to imagine that scenario. The treatment of Uyghurs is unjust and a clear violation of their human rights, but it's nowhere near the level of depravity of the western-backed Gaza Genocide.
> Like I said they have their own style of colonialism.
That's moving the goalposts and doesn't address the issue.
>They have been smart and done all their dealings via money.
You mean just like the country who issues the world reserve currency and who's intelligence agencies get involved in destabilizing regimes across the world?
> That's moving the goalposts and doesn't address the issue.
Is this how you make a constructive argument? Perhaps I was expecting too much from a joke account but this style of whataboutism is boring.
My post that you responded to set my premise which was that China has its own form of colonialism that is quite different than Americas but it exists and it’s quite strong. To classify China as a peaceful loving nation that respects other cultures is as if we were saying the US has never started a conflict. It’s factually a lie. China has a long list of human rights issues, they factually do not respect other cultures even within their own borders. I am not defending America but pointing out that China is not what the OP stated.
Are you the kind of superficial petty person who needs to take jabs at the messenger's name and not the message itself?
And are you really in the position to throw stones from a glass house with that account name? If you had your real name and social media profiles linked in the bio I'd understand, but you're just being hypocritical, petty and childish here with this 'gotcha'.
> To classify China as a peaceful loving nation that respects other cultures
I never made such a classification. You're building your own strammen to form a narrative you can attack but you're not saying anything useful the contradicts my PoV and wasting our time. Since you're obviously arguing in bad faith I won't converse with you further. Goodbye.
If you have an argument that is actually on topic with what I said please continue, otherwise save your troll account for someone else. The whataboutism/gaslighting is silly. You clearly cannot read threads or respond in a logical form to the right person. The conversation at hand was about China and in response to the OP classifying them as a loving and respectful nation. I made no attempt to defend the US and it has been you moving the goalposts. You throw about whataboutism around and then simply runoff with some flimsy excuse about multiple people being unable to converse with you. Troll account.
Cumpiler asked two very clear and direct questions:
>How many countries has China invaded and bombed in the last 30 years?
>How many deaths did China's warmongering caused abroad?
You didn't answer those, just started hand waving some stuff about China's "own form of colonialism" -- without even explaining what that is and how it works (which personally I'd be curious to hear about, and believe *is*" likely guilty of violence).
So you very clearly are the one guilty of shifting the goalposts, going on tangents, and bringing up usernames instead of real arguments.
I'm sympathetic to Infecto's positions, but I have to agree. I do think Cumpiler69's username is outrageous enough as to draw some commentary (provided it's civil and is semi-friendly ribbing) and perhaps even raise questions of whether they are a troll, but the substance of their comments is strong enough as to override these minor observations/objections.
the outcome would be exactly the same. AGI leads the human race off of a cliff, not in the direction of one human interest group vs another. the only difference would be that it was china that was responsible for the extinction if the human race rather than another country. i would prefer to die with dignity… the outcome we should all be advocating for is a global halt of AI research — not because it would be easy but because there is no other option.
we need to cooperate and put aside our petty politicking right now. the potential downsides of ‘racing’ without building a safety scaffold are catastrophic.
A serious question though, what does happen when AIs are filing lawsuits autonomously on behalf of the powerful, the courts clearly won't be able to cope unless you have AI powered courts too? None of how these monumental changes will work has been thought through at all, let's hope AI is smart enough to tell us what to do...
> A serious question though, what does happen when AIs are filing lawsuits autonomously on behalf of the powerful
It won't just be at the behalf of the powerful.
If lawyers are able to file 10x as many lawsuits per hour, the cost of filing a lawsuit is going to go down dramatically, and that's assuming a maximally-unfriendly regulatory environment where you still officially need a human lawyer in the loop.
This will enable people to e.g. use letters signed by an attorney at law, or even small claims court, as their customer support hotline, because that actually produces results in today.
Nobody is prepared for that. Not the companies, not the powerful, not the courts, nobody.
Unless you can afford your lawsuit to take up substantial time on Stargate and make a much stronger case than your average Joe who is still using o1 for their lawsuits
Oracle could reasonably be hit with some sort of stick every time they filed a frivolous lawsuit until the AI got tuned appropriately. Then it'd be a situation where Oracle were continuously suing people who don't follow the law, following a reasonably neutral and well calibrated standard that is probably going to end up as similar to an intelligent and well practised barrister. That would be acceptable. If people aren't meant to be following the law that is a problem for the legislators.
I'm envisioning a future where there's a centralized "legal exchange", much like the NYSE, where high speed machines file micro-ligation billions of times faster than any human can, which is decided equally quickly, an unrelenting back and forth buzz of lawsuits and payouts as every corporation wages constant automated legal battle. Small businesses are consumed in seconds, destroyed by the filing of a million computerized grievances while the major players end up in a sort of zero-sum stalemate, where money is constantly moving, but it never shifts the balance of power.
... has anyone ever written a book about this? If not, I think I'm gonna call dibs.
>A serious question though, what does happen when AIs are filing lawsuits autonomously on behalf of the powerful,
AI controlled cheap Chinese drones will start flying into their residencies carrying some trivial to make high explosives. With the class wars getting hotter in next few years we may be saying that Luigi Mangione had the right ideas towards the PMC, but he was underachiever.
I don't like how OpenAI turned majorly from what it was founded upon and their bias training ... but when considering the actual opponent here is China, it's not the worst.
I think OpenAI was originally founded against that kind of force. Autocratic governments becoming masters of AI.
I’m an American who was definitely raised in a “China Bad” world.
The last few months, between TikTok ban, RedNote, elections, United Healthcare CEO, etc I’ve seen so many people compare the US to China, and favor China. Which is of course crazy because China has things like forced labor and concentration camps of religious minorities, and far worse oppression than the US. But many people just view everything coming out of the US Gov’s mouth as bad.
Is the Chinese government worse than the US government? Probably. Do people universally think that still? Not really. The US Gov will have to contend with the reality that people -even citizens- are starting to view them and not their “enemy” as the “Bad Guys”.
I don’t get the good guys / bad guys mindset tbh.
Sure china gov is pretty bad and the us is by many metrics better - but why center your whole worldview around things that probably don’t affect you that much in your daily life?
US also has forced labour, huge prison population, bombing civilians and journalists to oblivion, literally nuking other countries and religious fanatics — do I still think china would be less pleasant as our new overlord? Yes — Do I think the world is better off with US-American hegemony? I’m not so sure.
Maybe it’s a net good for the world if not one power is dominating — maybe it’s the start of a hellish ww3.
I choose to believe the former.
Interestingly perhaps, as a foreigner (not from the USA or China) I can tell you not everyone around the world shares this perspective. There's people who trust China's single-party system over the USA's oligarchy.
By the time this project is done it will have been dead for 2 years.
Too many greedy mouths.
Too many corporations.
Too little oversight.
Too broad an objective.
Technology is moving too quickly for them to even guess at what to aim for.
Yeah, really the only thing missing from this initiative was the personal information of the vast majority of the United States population handed over on a silver platter.
That sentiment calls for reflection - whoever ends up on top of the heap after the AI craze settles down is going to be someone that everyone objects to. Elon Musk was himself an internet darling up until he became wealthy and entrenched.
That said, this does look like dreadful policy at the first headline. There is a lot of money going in to AI, adding more money from the US taxpayer is gratuitous. Although in the spirit of mixing praise and condemnation, if this is the worst policy out of Trump Admin II then it'll be the best US administration seen in my lifetime. Generally the low points are much lower.
Nietzsche wrote about these phenomena a long time ago in his Genealogy of Morality. there will never be someone who reaches the top who doesn’t become an object of ire in modern Western culture.
> That sentiment calls for reflection - whoever ends up on top of the heap after the AI craze settles down is going to be someone that everyone objects to.
I agree in principle. And realistically, there is no way Altman would not be part of this consortium, much as I dislike it. But rounding out the team with Ellison, Son and Abu Dhabi oil money in particular -- that makes for a profound statement, IMHO.
> That sentiment calls for reflection - whoever ends up on top of the heap after the AI craze settles down is going to be someone that everyone objects to. Elon Musk was himself an internet darling up until he became wealthy and entrenched.
Trying to process this but doesn’t his fall from grace have more to him increasing his real personality to the world? Sometime around calling that guy a pedo. Not much bothers me but at the very least his apparent lack of decision making calls into question many things.
> That sentiment calls for reflection - whoever ends up on top of the heap after the AI craze settles down is going to be someone that everyone objects to.
Did we see the same fallout from the space-race from a couple generations ago?
I don't think so — certainly not in the way you're framing it. So I guess I don't accept your proposition as a guarantee of what will happen.
A couple of generations ago we didn't have the internet and the only things people heard about were being managed. The big question was whether the media editors wanted to build someone up or tear them down.
The spoils of the space race would have gone to someone a lot like Musk. Or Ellison. Or Masayoshi Son. Or Sam Altman. Or the much worse old-moneyed types. The US space program was, famously, literally employing ex-Nazis. I doubt the beneficiaries of the money had particularly clean hands either
Elon Musk was an internet darling when his top character trait was "space! EVs!". Then he went Kanye/alt-right and weaponized twitter. It didn't have to do with the fact he has a lot of money.
Many people dislike all billionaires, but some have escaped criticism more than others by successfully appearing to have some humanity left in them, like Gates and Cuban.
This seems more like a move designed to frighten China -- or force them to spend money making LLMs -- then an actual threat. The clues are that Trump ceremonially blessed the deal but did not promise money (SoftBank et al will, supposedly), and then Musk said that's all fake because SoftBank doesn't have the money, and Altman countered that Musk should not be butthurt and should put America first. Who does that? I'm thinking, no one who has something real on his hands.
The way I think about effectful computation and IO in Haskell is as a kind of enforced metaprogramming. The actual language comprises a strongly typed meta-layer used to assemble an imperative program from opaque primitives. Like all powerful metaprogramming systems, this allows doing things you cannot easily do in a first-order imperative language. On the other hand, this makes all the 99% of simple stuff you don't have to even think about in an imperative language necessarily roundabout, and imposes a steep learning curve. Which contributes to the empirical fact that not many people are rushing to use Haskell, the superior imperative language, in real-world projects.
> The way I think about effectful computation and IO in Haskell is as a kind of enforced metaprogramming
If you think of effectful computation in Haskell to be a sort of metaprogramming, then indeed it will seem very complex, absurdly complex actually.
If you think of it as just the same as programming in any other language, except functions that do effects are tagged with a type that indicates that, then I think things will appear much simpler.
I prefer thinking of it as the latter, and it's been very effective (pun half intended) for me.
I do find people's faith in Democracy, as opposed to Authoritarianism, somewhat exasperating. Two candidates, pre-selected by the powers that be to lead the nation, compete in inane televised debates, wave flags and make promises that everyone knows they are going to break. This everyone debates hotly, and then lines up to register one bit of Holy Democratic Choice, to be averaged with a hundred million similar bits to determine, by a margin of a few percent, the one and only legitimate Government of the People, by the People, for the People. My Ass.
In the end, "democracy" is about power and control, just like any other form of government, and the TikTok ban is just another power-play, however it may be justified publicly. Not that I'm overly sorry to see it banned, by the way :)
Until very recently, "Democracy" was a dirty way to describe a government. It was in the same class of failed government models as tyranny, the rule of the mindless mob.
Maybe, but my point is that democracy is not even the rule of a mindless mob, more like mob rule theater. Ruling implies receiving information and performing complex actions and giving many and nontrivial orders. From a purely information-theoretical perspective, it requires a lot of entropy flowing from the decision maker to subordinates. On the other hand, national elections collect a tiny pool of entropy from the supposed root source of power and legitimacy, the people. This is not enough to rule a country, by many orders of magnitude. The country is instead ruled by ambitious individuals, who seize power in various ways - connections, backroom deals, backstabbing. Some participate in the election theater.
1. The countercurrent heat exchanger achieves exactly that: exhaust gases are cooled while the inflowing fuel mixture is heated up.
2. Thermophotovoltaics in general can operate with any heat source, though this device is clearly optimized for combustion. However, the efficiency is far too low to compete in the large-scale power generation segment. This is almost certainly aimed at light aviation, heavy drones, military applications, etc., where there are not a lot of alternatives that combine small size, high power density and good efficiency.
I suppose for aviation at least this is no less efficient than a gas turbine or a piston, and it's certainly a good deal quieter, has fewer moving parts, and requires less precision engineering than a jet engine. This feels tailor-made for attritable low->medium performance aviation, aka loitering munitions and drones. Strip away the "green" talk, and you're left with something that can burn just about anything (including hydrocarbons like avgas) without the complexity of a turbine.
maybe so. i don't know about attritable for the first applications though. may long range or duration oversight. a large % of the cost is these specialty cells which have not been scaled up to mass production. in the denominator is the intensity of light we can produce, which is based on how high a temperature we can drive, there's a very nonlinear brightness vs temperature. but at 100 suns or so we can get near to $1/W on the cells at startup scale
I can see that being a good use, ultra-quiet ISR that can stay aloft for extended times and doesn't require the complexity of a jet turbine? There has to be enormous demand for that.
Or turn it around a bit. If the entire device could operate at high pressure, then one could imagine putting it inside a jet or rocket engine. Feed it compressed fuel/air mix, burn, extract some energy via thermophotovoltaics, and blast the exhaust out a nozzle or use it to spin a turbine to drive a bypass fan.
An obvious down side is that most jets have very, very high fuel flow and power output, and the area required to extract enough electricity to make this whole exercise worthwhile may be excessive. Also, a lot of military applications are not going to like that sodium illuminant lighting up the exhaust gasses, scattering radar, or otherwise making the plane more visible.
edit: I see that there’s an effort to recirculate the sodium. Maybe that’s enough.
Fuel is burned to head sodium, if you are getting too much heat for your taste you can burn less fuel. It is kinda the goal of the exercise.
But in any case, I believe that the more you heat sodium, the more light it emits, probably there is a practical limit on an incoming heat power after which the thing will go boom, but before that it will follow some roughly linear law: the more heat energy in, the more light comes out. Though I'm not a physicist, so I make be wrong, even if I do not see how I can be wrong.
1. https://orcutt.net/weblog/wp-content/uploads/2015/08/The-Tro...
reply