TeknOShEeP
23hr ago
animal brains in general are extraordinarily skilled at learning what's necessary for success in their environments.
Yes, animal brains are extrodinary examples of specialized, niche intelligence. What very clearly sets homo sapiens apart is general intelligence, and the ability to learn things devoid of instinctual context.
Even if you take Pascal's wager seriously, it is not actually very useful.
I find it as useful as AI safetyism- a highly uncertain system with allegedly vast consequences easily subverted by defectors. Fortunately I'm very much on Team AI Fizzle, so the fact that its clearly unworkable nonsense doesnt matter to me.
they did the math which showed that a chain reaction was impossible prior to the test.
No, they did the math to suggest that based on what they knew at the time, a chain reaction was unlikely to a fairly high, but very much not unity, degree of certainty. And as the Castle Bravo whoopsie showed a few years later, that same group of very smart people was capable of making some fundamental errors resulting in catastrophic consequences.
TeknOShEeP
1d ago
Uh, object is a singular noun. You are describing collective and/or plural nouns. It is trivially true that they are more complicated in the multiple, but in no way renders the claim incorrect.
TeknOShEeP
2d ago
But the fact is that the possibility is there, and is significant enough that it currently cannot be dismissed out of hand. Thus it makes sense to halt development until we are certain that this research won't doom us all.
I find it ironic that this is the logic used by a group that pretty much universally rejects Pascal's Wager. Also, it wouldn't be the first time humanity has made this particular calculation- when the first atomic bomb was tested at Trinity, Oppenheimer was "pretty sure" it wouldn't cause a neutron chain reaction and ignite the atmosphere in a nuclear hellstorm, but he couldn't guarantee it. Infinite stakes do not necessarily require infinite caution.
TeknOShEeP
2d ago
doesn't assume that -- it rests solely on the idea that brains are physical objects. This is empirically verified by every single experiment run on a human brain. More generally, it's been borne out on every noun that interacts with the physical world.
This is missing the part where the human brain is an exceptionally well-tuned physical object shaped by millenia of evolutionary pressures that arguably constitute a training set vastly bigger than the laws of physics as we currently understand them say is possible to match with an artifical model, much less do any meaningful computation with.
It is also missing the part where the human brain is the most complicated object in the universe, as it is the only currently known object capable of of understanding these questions well enough to even ask them. And even it does not fully understand itself.
TeknOShEeP
2d ago
I dont think its uncharitable, from my reading of various AI safety writings I would describe it as substantially accurate.
TeknOShEeP
2d ago
Nah, not even that. The NNPT is just about the big kids deciding the big stick club should be exclusive to them only. And its enforced by big sticks, not some figleaf of global regulation. It would only work if AGI already existed but was a closely held secret of the militaries of the various signatories.
TeknOShEeP
2d ago
IIRC from the last time Nate Silver brought it up, historically Generic Republican underperforms by roughly 4 points on congressional polls, while Generic Dem overperforms by 3 for a relatively constant delta. Modern polling (ie after 2016) has apparently slightly closed this gap, but it still exists.
TeknOShEeP
3d ago
Generic Dem outpolls basically any Dem candidate in any election. Its a fabulous example of just how good Dem branding is that their voters care so deeply about the letter next to the name on the ballot. Its only when an actual human with scandals and bad interviews gets on the ballot that Dems start losing. Republicans have the opposite problem- Generic Repub looses polls badly.
TeknOShEeP
4d ago
I think it is highly unlikely that Russia uses nuclear weapons on any targets. That being said, if Russia were to use nukes, it would almost certainly be on European targets, with Ukraine at the top of the list. Aside from the tactical considerations, from a threat management perspective there are a) only two European countries with credible nuclear deterrants at all, and frankly neither of them are particularly impressive (France- 4 SSBNs, 1 at sea at all times, and a couple of airbases with Rafales capable of air-launched nuclear cruise missiles; UK- 4 SSBNs, 1 at sea at all times; 4 well-timed strikes could theoretically eliminate all available European deterrance) b) its a lot easier and a lot more immediately valuable to hit targets in Europe if Putin does decide to sling nukes.
That being said, I'm pretty sure this would be a suicide ploy, as the US would likely decide the world order cannot continue with Putin and cronies in charge, and remove him by any means necessary- including ICBM or cruise missile strikes. In a mutual nuclear exchange, Russia loses very badly, possibly to the point of landing zero-one hits on the continental US while suffering functional annihilation in return.
TeknOShEeP
6d ago
Chinese and Indians have completely invaded American STEM at all levels: tech companies, academia, mathematical finance, hardware, and engineering.
Actual engineering (not software "engineering") is still very white in the US. Especially things like aerospace, civil, and nuclear engineering. Immigrants tend towards the higher pay/effort fields, and ones that aren't gatekept by ITAR.
TeknOShEeP
14d ago
Jets fans
Yes, but consider this: Jets fans are all deviants who get off on hating things they love.
TeknOShEeP
26d ago
Yes, the S-1 is a marketing document disguised as a regulatory filing. If it was 5 years ago, it would be full of crypto and NFT bullshit. As it stands, those AI valuation numbers are ludicrously overvalued, as they will be for any of the other upcoming AI IPOs. (I do not hold SPCX for this reason, when it drops to reasonable levels I will buy).
Conversely, what is ludicrously undervalued is the TAM for space services. $370 billion is a third of the annual aerospace market, and not that far off of what is actually spent on space services today (roughly $250 billion). Its downright prude compared to the AI hype machine, and given SpaceX's rather large lead in that sector is the area I would focus on most as an investor.
TeknOShEeP
26d ago
its only downside being that its market offers (relatively) limited upside.
On the contrary, the upside for space stuff is functionaly infinite. Its just been handicapped by extreme launch costs for 50 years. Every financial analysis writing on SpaceX gets this very wrong- the market for launch services grows in a highly non-linear fashion, inverse to launch costs. SpaceX is legitimately poised to chop two orders of magnitude off $/kg-to-orbit, but this gets priced in as like "10-20% annual growth". Its like pricing in a similar number for the internet in 1992, because data connectivity is still a fairly expensive thing.
TeknOShEeP
26d ago
I have a hard time squaring this dual narrative. Are they a space company, or an AI company?
Both. AI datacenters in space give an investor thesis to why they deserve the valuation. Because right now, AI sells. That being said, I think the S-1 dramatically oversells AI TAM, as will all AI-related IPOs in the near-term future, but dramatically undersells the space side of the business. My prediction is that in another 20 years, SpaceX will be a space-services conglomerate that happens to have an AI division, rather than the flipped version given in the S-1.
TeknOShEeP
1mo ago
Hah, funnily enough thats the other model we had under consideration. We test drove both the base gas and premium phev versions and absolutely hated it. $34k stuck us as about $14k overpriced. Might as well just get a used lexus or something, and be much better off.
I think the cheap new car market right now is in a bad place- you can get fairly heavily depreciated but still new-ish and reliable upmarket vehicles for the same price and a much better ownership experience.
We'd both much rather buy a nice new car and drive it for decades (our newest current vehicle is 11 years old but still comfortable and reliable) than get a cheaper one that we'll get tired of and have issues.
TeknOShEeP
1mo ago
The prepayment wouldnt be an issue as the whole point is to stretch it out as much as possible. And theres no back interest, its simply a loan at 0%, if its in arrears it goes to collections/repo. Our intent is to have our finance guy create a separate account that we put $60k in and have monthy payments pulled out of, with the remaining principle invested. Naturally theres risk of losing principle, but thats true of all investing. The exact investment profile is TBD, i was just using S&P for some back of envelope calcs.
TeknOShEeP
1mo ago
Okay, need some double-checking of my thinking. With an expanding family, and expanding home needs my wife an I are getting a bigger (and safer) vehicle. Long story short, its $60k, which we can actually cover outright without strain. If we do, there is a manufacturer incentive in a $5k cash rebate, which is quite nice! However, a separate, mutually exclusive incentive offer is $0 down (though we will have to pay roughly $6k in taxes and fees upfront, but that happens no matter what option we chose), 0% APR for 72 months financing, which seems to me to be an absolutely screaming deal. Some quick math suggest this saves $12k over a conventional loan, but even more importantly if we just throw that $60k into an index fund and pull our loan payments out of it, we will still have between $22-26k left at the end assuming average returns. We tend to keep our cars a long time and don't envision wanting to change out of this one. Is there any good reason not to do this?
Its detailed in the linked article, but what the SPLC actually "sells" to its donors is the ability to do a massive, coordinated pressure/harrassment (depending on your outlook) campaign against noncompliant institutions that results in reputational damage that can be priced very highly.
So in some sense, it doesnt even matter if the SPLC is convicted, the mere fact that they now have the "potentially fraudulent" label hanging on them means compliance departments not only do not have to, but are arguable legally barred from, having contact with them in the way they used to operate. So the SPLC as a political operative NGO is basically dead in the water.
Also, not sure how they are going to weasel out of their CEO directly admitting to bank fraud in official communications, so as a going concern as well they seem to be in a lot of trouble.
TeknOShEeP
2mo ago
Nah, thats a "DEI admit" flag if ever I saw one. Caltech used to be the last holdout against it, but even they have collapsed in recent years. I would bet his grades in undergrad are of the "barely passing with massive help from tutors and the counseling office" vintage.
TeknOShEeP
2mo ago
TeknOShEeP
2mo ago
Louisana, Texas, Florida, Indiana. Contrast with Connecticut, California, and now Virginia. It seems like math disagrees with your conception of reality.
TeknOShEeP
2mo ago
In the case of the SPLC, they are the same thing. The organization only exists as long as donors can be convinced the big scary evil hate organizations actually exist in any meaningful way. When the SPLC was founded, the KKK was an actual social force that had real political heft in the South. They are now a joke in internet memes. The biggest "racist gathering" in recent memory, Charlottesville's "Unite the Right" appears to be at least partially funded and organized by the SPLC themselves. There are functionally zero active, violent, White race groups in the US, and thats a big problem for the SPLC, because thats their entire raison d'etre. Its not that demand outstrips the supply, its that the supply is entirely fictional.
TeknOShEeP
2mo ago
The US actually throws politicians in jail fairly regularly for corruption and bribery. Not often enough for my tastes, but it happens. Just yesterday a congress critter resigned minutes before being expelled by a bipartisan vote, and the federal charges are probably going to stick.
The informant the SPLC paid 7 figures for was fundraising for the most milquetoast-ass Nazi group possible who seem to be decades removed from any convictions for violent crimes. Safe? No, but La Cosa Nostra or the cartels they are not.
It hasn't yet, but as recent events show, it is enforced through stealth bombers dropping 30,000 lbs bunker busters and massive air raids. Or full scale ground invasions deposing the existing governement. All of which I would classify under "Big Stick" diplomacy.
More options
Context Copy link