I don’t really think the issue is that fact that ChatGPT (or perhaps another llm) was used.
The issue is that it was not checked. Perhaps they used it just to translate as a final step and did not have a fluent person to check it over (with the requisite level of understanding) LLMs are new, and soon to be superseded by multimodal AI, so the understanding of them is currently quite immature, in both usage and consumption of.
I'm very bullish on fusion given the new magnet breakthroughs, detailed here: https://www.youtube.com/watch?v=KkpqA8yG9T4 The SPARC reactor is still on-schedule to be completed and operational in 2025. It's expected to generate net-positive electricity, on the order of 140MW pulses. A larger version to supply energy to the grid is planned for the 2030's.
My concerns have moved to "okay, what is the cost of things even if energy is 'free'?". For example, desalination -- currently the majority of the cost of desalination is in the energy used, but even if the energy is free, the materials cost and capital expense is currently about 40% of the cost.
Will the cost of the inputs also go down dramatically as a result of fusion? Or will desalination still be cost prohibitive for most developing areas due to materials? If desalination price falls 99.9% -- what environmental costs will be associated with the extraction/transportation of necessary materials as well as the warm & highly saline effluent that will be dumped into surface waters?
With unlimited clean energy, some ridiculously inefficient schemes can become practical, such as removing CO2 from the atmosphere.
Well it's no joke, all the money chasing the mirage of D-T fusion for energy production; even if it works it will not be any kind of panacea. Other fusion types might be more realistic.
Imagine you are autocorrect, trying to find the most "correct sounding" answer to a the question "Who won the super bowl?"
What sounds more "correct" (i.e. what matches your training data better):
A: "Sorry, I can't answer that because that event has not happened yet."
B: "Team X won with Y points on the Nth of February 2023"
Probably B.
Which is one major problem with these models. They're great at repeating common patterns and updating those patterns with correct info. But not so great if you ask a question that has a common response pattern, but the true answer to your question does not follow that pattern.
Yes, it actually sometimes gives C and also sometimes B and sometimes makes up E. That's how probability works, and that's not helpful when you want to look up an occurrence of an event in physical space (Quantum mechanics aside :D).
I've never had it say 'I don't know', but it apologizes and admits it was wrong plenty.
Sometimes it comes up with a better, acceptably correct answer after that, sometimes it invents some new nonsense and apologizes again if you point out the contradictions, and often it just repeats the same nonsense in different words.
one of the things its exceptionally well trained at is saying that certain scenarios you ask it about are unknowable, impossible or fictional
Generally, for example, it will answer a question about a future dated event with "I am sorry but xxx has not happened yet. As a language model, I do not have the ability to predict future events" so I'm surprised it gets caught on Super Bowl examples which must be closer to its test set than most future questions people come up with
It's also surprisingly good at declining to answer completely novel trick questions like "when did Magellan circumnavigate my living room" or "explain how the combination of bad weather and woolly mammoths defeated Operation Barbarossa during the Last Age" and even explaining why: clearly it's been trained to the extent it categorises things temporally, spots mismatches (and weighs the temporal mismatch as more significant than conceptual overlaps like circumnavigation and cold weather), and even explains why the scenario is impossible. (Though some of its explanations for why things are fictional is a bit suspect: think most cavalry commanders in history would disagrees with the assessment that "Additionally, it is not possible for animals, regardless of their size or strength, to play a role in defeating military invasions or battle"!)
on some topic at least it correctly identify bogus questions. I extensively tried to ask abount non existent apollo missions for example, including Apollo 3.3141952, Apollo -1, Apollo 68, and loaded question like when Apollo 7 landed on the moon, and was correctly pointing out impossible combinations. this is a well researched topic tho.
Only if it’s a likely response or if it’s a canned response. Remember that ChatGPT is a statistical model that attempts to determine the most likely response following a given prompt.
My point is he could buy a lot of time to save the business. He's not "running out of time". The entire article just feels like: woe is the billionaire's pet project failing because it's not profitable in an unreasonably short amount of time despite plenty of cash availability if said billionaire wants to continue and achieve their original goals.
Where did anybody expect Blue Origin, SpaceX and Virgin Galactic by this point? 17-20 years of R&D, building on decades of rocket science and space exploration, consulting with NASA et al. and between the three of them, only 1 has been able to send humans into orbit and that was only very recent.
20 years to go from a pile of cash and no experience, to a full production line of safe orbit-capable launch vehicles is a very big ask. SpaceX has achieved that, but at the cost of many of its employees (just look at the press around SpaceX working conditions).
Yes, it's longer than the time between the Mercury missions and Apollo 11, but the 50s-60s era of space travel was significantly higher risk. If corners could be cut in the same way, no doubt 20 years would have been fairly reasonable. Add in an existing presence in space R&D and the shorter time frames of the space shuttle program seem even more reasonable. But that just isn't possible in today's safety first world.
Well luckily we have markets that can quickly turn equity in Amazon into cash. Bezos has in fact been doing that to the tune of billions of dollars a year to fund Blue Origin. I don't see him stopping any time soon.
1) time and time again, particularly in the media, I hear net worth bandied about as though it's the same as cash. Bezos is worth X billion - think what that cash would pay for!
2) billions in equity can't be translated into billions in cash. Millions, for sure. But not billions.
3) any statements that talk about taxing the rich inevitably seem to come down to transferring equity to the state over time based on a current stock market valuation.
All of the above seem misinformed and/or a bad idea, and worth trying to establish fundamental understanding in this area. If you already have that understanding then that's fantastic.
Jeff can define success however he wants. He can define it in terms of making money, or in terms of successful launches. Making money is hard - there is only so much demand for things in space, and some competition - making it questionable if the investment will pay off. However launching isn't as hard, and he can define metrics in that way and be successful for his purposes.
How is launching things into space any different from paying composers to write much, or painters to paint pictures - two things that rich people regularly did 300 years ago.
Space demand will grow incredibly fast very soon as SpaceX leads the incredible prices drops in $ per KG to orbit.
I'd love for them to be successful, and its great to see so many space companies being created, testing, and launching. And it's very important for the future of civilisation. That doesn't mean that you can't ask questions about the health of a particular company.
He needs to get something in orbit, anything. Once he has a single win, he can start scoring launch contracts and competing in the launch provider market. He should end up more profitable than every provider with disposable launchers.
New Glenn should be that vehicle, if they ever finish it.
> By definition trade in capitalism is done willing.
Not by the definition used by the people who named and defined capitalism.
It's true that after that, the conceit that capitalism involved only voluntary, uncoerced trade was adopted by it's defenders as a rationalization of the system, but that was not true of either the specific real world systems for which the name “capitalism” was coined to refer or subsequent real world examples, and certainly has nothing to do with the definition of capitalism.
If you want to distinguish the proposed scenario from capitalism, it would be in that it does not involve private property rights in the means of production, but instead on their forcible seizure and defense, but that's a slippery distinction because commonly such systems evolve into a degree of legitimization and trade with recognized rights between the parties, and the roots of capitalist property also start in forcible seizure which is later legitimized.
> If someone living today called themselves a capitalist would you expect them to be involved in "forcible seizure and defence"?
Given the diversification most capitalists have and looking at what major corporations do globally, yes, though I'd also expect them not to think of themselves that way.
Drug cartel leaders, I'm sure, often have similar self-serving rationalizations of their role.
From related portions of the economy: given the growing use of debtor's prisons, predatory loans, and coercive tactics including armed repossession & bounty hunting dependent on an exploitive for-profit bond regime? Yes. These are not companies rejected by modern capitalism. On a higher economic level private equity's leveraged buyouts are very frequently hostile takeovers that use a company's own resources to seize control of it.
> By definition trade in capitalism is done willing.
The definition of capitalism, and the world in which capitalism operates, are different.
The final transaction between buyer and seller is voluntary.
But all of the backend infrastructure may be highly manipulated in unethical, forceful ways.
A person buying some whale meat willingly pays the merchant at the meat market.
But that whale meat was acquired because one group killed the whale before another group. And that group killed the whale first because they setup groups who threatened other would-be whale hunters, and as this group gained a bit of financial traction they paid off local officials to pass some “coastal safety” ordinances that provide them some level of monopoly on killing whales, and worked out another ordinance that lets them dump toxic byproduct in a local river to place some of their cost into the public that won’t be easily rectified for decades.
So a perfectly ethical capitalist fisherman might well find themselves facing men with guns who forcefully prevent them from competing, when the police show up to enforce the local coastal safety law.
