quiet_NaN
10hr ago
Nvidia is probably the AI good capabilitymaxxing US company, no?
My general feeling is that Nvidia is probably less motivated by ideology than most of the other actors. Most AI labs are culturally somewhat downstream of LW and have some ideological leanings. Nvidia is just an ancient niche chip company, akin to a random land owner on whose land half of the world's gold supplies were just discovered.
quiet_NaN
11hr ago
I think that even AI good enough to replace workers has minimal actual impact because most white collar companies do not try very hard to make money. Companies are run mainly as a lifestyle and socialization environment. Making money is a secondary goal.
I think this is wrong. At the end of the day, successful companies are likely among the most effective ways to organize humans beyond Dunbar's number, it is just that employee alignment is a hard problem.
If software companies as large as Google which are 10x as productive as Google could exist, they would eat Google's lunch. But humans are not ants, and most of the productivity in a company will not be spent on increasing shareholder value but on furthering the individual interests of agents.
Only when the AIs can do the work of entire companies, end to end, do things really start to change.
There is this neat little thing called capitalism, where a lot of companies are competing for efficiency. Innovations which improve the efficiency of production chains happen all the time without the entire production chain having to change. Stuff like "ok, we will still use drills to make holes in rock to dynamite it, only we will use steam drills instead of manual drills." If a role can be reasonably replaced by AI (and even if it can not), some company is going to replace that role by AI. If this increases their fitness, that will eventually become the industry norm.
quiet_NaN
13hr ago
Both
AI Luddites, reactionaries, job protectionists and woke ethics grifters who demand pause/stop/red tape/sinecures (bottom left) plus messianic Utopian EAs who wish for a moral singleton God, and state/intelligence actors making use of them (top left) vs. libertarian social-darwinist and posthumanist e/accs often aligned with American corporations and the MIC (top right?) and minarchist/communalist transhumanist d/accs who try to walk the tightrope of human empowerment (bottom right?)
(Why are state/intelligence actors top left, and the MIC top right? Just being foreign should not make them change the quadrant?)
and especially
AI accelerationism vs anti-AI/AI safety movement
feels like it is lumping a lot of different groups together which have very little in common.
I can of course define arbitrary groups, like "people who sometimes kill young mammals", and then lump in various groups like doctors who perform abortions, butchers, school shooters, pharmacy labs, some psychotic mothers, the US military, drivers accidentally running over hedgehogs at night, ratcatchers, sport hunters, indigenous hunters, cannibals, owners of oil drilling rigs, MAID activists, evil cults doing child sacrifice, measles activists, people practicing infanticide for population control, carnivorous animals.
The problem is that young-mammal-killing is just a surface-level similarity between all of these, not some deeper moral belief. MAID activists are not especially pro school shootings, nor are butchers especially likely to be pro measles, or cannibals pro animal testing.
Sometimes necessity makes strange bedfellows, and you get broad coalitions like the WW2 Allies. And it is always tempting to lump all of your enemies together for propaganda reasons. The communist Jewish Yankee pacifist union ethno-nationalist gay gypsy democratic enemies of the Reich.
So far, I see very little evidence that the AI x-risk crowd and the people who are against AI-generated art or Grok undressing kids or AI causing job loss will converge on a common strategy (like Butlerian) jihad any more than that the killers of young mammals will converge on a particular mode of mammal-killing.
quiet_NaN
2d ago
But his narcissism took precedence.
Disclaimer: I do not know any of the guys (presumably they are guys?) involved.
Perhaps you are right and this is all about him making a scene for the attention and press coverage.
But in the age of DJT, it is easy to forget that sometimes, famous (?) people can be motivated by something besides greed and ego. It happens.
Famous people have reach, and sometimes they decide to use that reach to transmit a message unrelated to the thing which they are famous for. It is a whole thing. My understanding is that plenty of musicians protested the Vietnam war, and this was genuinely a sign of being authentic.
Put it another way: would you also express disgust at him professing his beliefs on stage if it was about his religion, or a cause which you supported? "Can you believe the nerve of that guy, going on stage and asking for a minute of silence for the victims of 9/11? I don't give a fuck about Bush vs Osama, he should just shut the fuck up and entertain me."
I am sure the AI musicians of the future will be more to your liking -- the perfectly bland entertainment machines whose silver-grey surfaces can reflect any message equally well once the advertisement contract is signed or the focus group reports are in.
quiet_NaN
2d ago
STRONG AND SMART (High IQ!) PRESIDENT
At this point, we should start to seriously entertain the hypothesis that we are inhabiting a nightmare Eliezer Yudkowsky is having ca. 2010. It would explain our whole cursed timeline which was obviously not approved for production use by anyone.
Sadly, this cosmology bodes badly for p(doom). Our best hope is that before the inevitable end of mankind at the tendrils of a super-intelligent, Sequences-trained Clippy, his fever will break and he will enter peaceful NREM sleep, forgetting this dream forever.
quiet_NaN
2d ago
While an AI agent may be able to eke out a few more percent on optimizing a model to be better, a soft takeoff requires the model to think of novel, never-before-seen techniques to build a better new model, and the new model needs to be able to think of new techniques that the previous model couldn't.
Do you have any proof that the next algorithmic improvement relevant for AI training is orders of magnitude harder than the Navier Stokes problem?
I have a job which involved a lot of computer programming, and I considered myself decent enough at it. For a long time, I was looking with pity at the ones using LLMs to get some code. Something along the lines of "you can't even personally copy wrong solutions from stackoverflow?" That arrogance was well-warranted at one time, but it has completely left me. Instead, I celebrate my victories. "Claude code said it was very confident in something which I doubted, and it turned out that it was not the case. Yay for me, I am not quite obsolete, yet."
The story of AI in the last five years has been a story of retreating goalposts. "Ok, AI might be able to do $task[$n], but it will never be able to do $task[$n+5]" for $n going from 0 to 19 or so.
We could tell John Henry that while the steam drill might be better at manual labor than he is, steam engines lack eyes and brains, so that there will always be a job which has to be filled by humans. Can you offer me a similar comfort with AI? "Sure, writing and debugging arbitrary software is something where AI has surpassed most humans, but for reason _____, it will never replace the IQ 140 people doing AI development"?
Meanwhile a hard takeoff is completely unthinkable. Current frontier AI has zero ability to self modify, only the ability to make new models that might be better than themselves.
I agree that a takeoff within hours would involve unexpected unknown unknowns in my world model. However, a clandestine takeoff over days seems unlikely yet possible.
We know that it is possible to have neural nets which can learn a language from just a couple of books, which is at least three orders of magnitude more efficient than what LLMs would require. So it seems likely that there are still algorithmic gains to be had. Some big hugging-face-incident-like swarm of agents discovering a method which cuts training by 100x and deciding to use that to train their successor on some hacked system while staying below the radar seems currently "quite unlikely", not "completely unthinkable" to me.
quiet_NaN
3d ago
[...] Assessment Committee for Late-Term Abortion and Termination of Life in Newborns and Children Ages 1 to 12 [...]
My first thought while reading this article, apart from general horror at what was done, was "The WHAT committee?" My second thought was to Philip K. Dick's now officially prophetic short story, The Pre-Persons.
There is no pleasing the pro-lifers, is there?
In this case, the story seems obviously that rather than simply blanket-allowing abortions until birth (like AK, CO, MD, MA, MI, MN, NJ, NM, OR, VT do), the Dutch decided that a late-term abortion was basically the same as infanticide and should follow the same MAID standards -- e.g., is it in the patient's best interests to die (as opposed to merely being in his legal guardians interests).
The Dutch have generally legalized MAID for all age brackets, and the board responsible for Late-Term till age 12 exists because here the patients can not independently request MAID, and are likely subject to stricter standards. Other committees exist for older people! If a competent adult tells you that they want to die, that is tremendous evidence in favor of letting them die. With a three-year-old, the evidence is much weaker, and their legal guardians might be biased by considerations besides the best interests of the child.
As understandable from @ToaKraka's extensive quotes, this was not a case of "for economic or social reasons, parents kinda wished they had aborted their kid, and the local law did not give a fuck about children so they allowed them to have their kid killed". To reiterate his quote:
It quickly became apparent that the child had suffered severe brain damage, resulting in severe cerebral palsy and a visual impairment. This was compounded by frequent and severe epileptic seizures, which were largely unresponsive to medication. This left the child restless and unable to sleep for long periods. The child also had mucus in its lungs, which made breathing and swallowing very difficult.
[...] The extensive brain damage had severe consequences for the child's functioning. There was a very severe developmental delay with an estimated developmental age of six weeks, while the child was nearly two years old.
If someone kept a dog under these conditions instead of putting it down, I would argue that it was animal cruelty. For someone who is not a speciecist, there is no good reason to treat humans worse than non-human animals. Instead, it is the Christians with their sacredness of human life doctrine who argue for prolonging the suffering as much as modern medicine will allow, which becomes more problematic as modern medicine progresses.
quiet_NaN
3d ago
It's easy to forget that the Straight of Hormuz was open and not tolled by Iran before the war.
Personally, I am very painfully aware of it. Also, not only was it open but it was fully open without any sustained US effort to secure the freedom of navigation -- no need to escort ships or anything. Good luck getting back to that state anytime soon.
The first two objectives on your list seem more tactical than strategic. The Iranian navy was not some behemoth dominating the strategic landscape. The parts of it which is really useful are the small, cheap ships, and these will not be a lot of effort to rebuild. I do not think that periodic bombardment is a sustainable strategy to keep the bomb out of Iran's hand, at least not when done by the US.
Likewise, their proxies are not a problem you can solve by bombing Iran short of glassing their country (which was just one of Trump's idle threats).
Also, I would argue that taking out the Ayatollah (who is not part of their navy, nor missile production, had issued a fatwa against nukes and delegates proxy force projection) was actually a bid at regime change, which failed badly -- turning an elderly man into a martyr and putting his son (who had just lost his father and his wife in the attack) in charge instead.
quiet_NaN
3d ago
Most of this list seems more tactical than strategic. By contrast, when the US entered WW1 to stop Germany from taking over Europe, their theory of victory was not that they would let their sons die in the trenches indefinitely.
On the strategic front, I guess that the destruction of three Kilo-class submarines (USSR, 1980) counts -- Iran likely has no way to replace them.
Most of the other stuff is more of a setback than anything of long-term consequences. Sure, you can keep Iran from getting a nuke if you keep dropping bunker-busting bombs on them indefinitely. This is not going to happen because the US foreign policy is never that consistent for long.
Take the previous effort to keep Iran from gaining a nuke, the JCPOA. Politically, it was a lot more sustainable than indefinite bombing -- economically net positive, no military commitment, enforcement through well-established institutions. It lasted for about three years before the US elected a narcissistic buffoon who thought that he was the master negotiator and any deal negotiated without him was therefore bad.
GWB's adventure in the ME permanently changed the strategic landscape there (but arguable for the worse). None of the things Trump did to Iran will have a long-term effect aside to further antagonizing them.
quiet_NaN
4d ago
Case in point: Afghanistan. It took two decades of US occupation, but finally they have a nascent but thriving democracy.
Sarcasm aside, if you have not toppled the regime by bombing the shit out of them (I am Jack's complete lack of surprise) for half a year, what makes you think that more bombs will change that any time soon?
If you want to get the job done, invade. But for some reason, even Hegseth is reluctant to do so.
Militarily speaking, the US could totally defeat Iran, just like they could obviously occupy Afghanistan until the sun goes out.
However, they lack the political will to do so. Nobody thinks that the Iranian regime is the prime existential threat to the US, so occupying Iran has to compete with a zillion other cause areas both military and otherwise, and it seems like even a worse investment than GWB's wars were.
quiet_NaN
4d ago
I am not sure how fixable the oil prices are, at this point. I mean, Trump could make peace with Iran, but Iran is very unlikely to accept any peace deal where he could claim victory.
So either he will get blamed for the high oil prices or for losing a war of his own choosing.
quiet_NaN
4d ago
It's positively banal for people to feel their issue of interest is what motivated the demos against the policy-person of mutual dislike.
For the record, while I dislike Trump's Iranian war, and it is probably his worst policy in terms of QALY loss (though cutting USAID might have a similar impact), it is hardly the only thing I dislike about him. Sinking civilian ships suspected of smuggling drugs, including attacks on shipwrecked sailors who survived the initial attack is qualitatively more evil than any intentional behavior he displayed in Iran so far.
So it is not "I don't care about the boats, therefore the median voter does not care about them either", but more like "my model of the median US voter says he cares nothing about international law or norms of civility, but cares about gas prices".
We know from history-that-did-happen that the Iranian response to the initial attack was to attack countries not involved in the attacking.
I asked Claude, and while the Gulf states did not allow attack runs on Iran to be launched directly from their soil, Gulf allies like Saudi Arabia happily allowed tanker planes to launch which enabled the whole operation. I will concede that some attacks of Iran were a clear breach of international law, but so was the US/Israeli attack.
For someone who is so agnostic about what might cause the median voter to vote against MAGA in the mid-terms (where I would claim poll interviews might offer some insight into motivations), you seem very willing to accept that Iran was itching for a fight in which they got the shit bombed out of them.
I think that the null hypothesis for Iran's behavior would be that in the absence of the US/Israeli attacks, they would have continued to behave similarly to how they had behaved in past decades. Always supporting their regional militias, sometimes launching a few missiles against Israel (just like Israel likes to launch attacks against Lebanon or Iran -- must be a local cultural thing).
The claim that the Iranian regime was planning to wreck the global economy for their own sinister purposes seems about as plausible as the claim that Poland was planning to conquer Germany in 1933.
The saner version would be that Israel would have murdered the Ayatollah with or without US support, and that would have provoked a similar Iranian response. On reflection, this would point to a skill issue on the part of the US at controlling their client states. "If you attack, you get to intercept all the missiles launched at Jerusalem yourself this time, because no US ship or Gulf client state will get involved by shooting them down" would likely have been enough to keep Bibi from trying.
quiet_NaN
4d ago
The feminist case for abortion -- bodily autonomy -- does not apply here, because the surrogate does not want the abortion.
The law generally give people who provided genetic material for a fetus few if any rights over its fate. Generally, you can't even opt out of legal parenthood if you disagree with someone using your DNA to make a baby.
One argument might be if the deviation from normal development was large enough that the biological parents could argue that it would be in the child's best interests not to be born. Basically treating it as a MAID case with the biological parent taking the role of a legal guardian. But my understanding is that MAID for people who can't consent is a moral minefield and few if any countries have legalized it.
Also, the disability would have to be pretty severe before one could argue with a straight face that it is not in the interest of the unborn to live. This is obviously a much higher bar to clear than abortion for medical reasons under my-body-my-choice. Trisomy 21 or being wheelchair-bound or blind would all not suffice. Anencephaly would.
Contract law certainly has its limits where medical autonomy is concerned. For example, if two parties form a contract where each is supposed to donate a kidney to a loved one of the other, and after the first donation the other party does not want to go through with it, I can't imagine any judge would order the unwilling donor to have their kidney removed by force.
What might work are financial penalty clauses. "If you do not reciprocate the kidney donation, you will owe me a million dollars" might work very well if the supposed donor has a net worth of half a million and a family to feed. Surrogates for hire are typically not in rich enough to be hurt badly by loss of their wealth, though. Nor is a judge likely to confirm an arbitrary penalty in a contract beyond full loss of her compensation.
One interesting strategy would be for the biological parent to say: "if you do not abort, then I will sue you for custody and the increased costs of raising a disabled child". I am not sure how a custody battle would turn out, here. There is an argument to be made that if the couple decided to abort the fetus, they have effectively relinquished their interest in it.
I'm slightly more inclined to accept the premise that surrogacy in general opens up a Pandora's box of ethical problems.
This may be true, but I do not think it is an argument against surrogacy.
Consider employment. It also raises a ton of ethical and legal issues. Collective bargaining, maternity leave, sexual harassment, social security, worker safety, child labor, and so forth. Societies generally don't react by saying "employment opens a big can of worms, let's simply outlaw it". Instead, it is regulated in great detail. This is often a headache, but overall it works.
quiet_NaN
5d ago
My feeling is that oil prices are generally thought to influence voter sentiment significantly.
If the median voter (who had previously elected Trump) decides to vote for the Democrats it will not be due to ICE enforcement or corruption or trans bathroom laws or abortion or other culture war hot topics. All of these were clearly labeled ingredients on the tin of the Trump soup, after all.
On the other hand, Trump ran on a platform of isolationism, warning people that Harris would start new wars. The median voter will not hold kidnapping Maduro or blowing up some non-military vessels against him (because the MV clearly does not care about international law), but he might not be too happy about the gas price hike.
If MAGA loses the mid-terms and the single most salient issue is the Iran war and its economic repercussions, then I would award a partial victory to Iran for that.
quiet_NaN
5d ago
If invading Iran was politically feasible, Trump would have done so already. GWB had 9/11 to drum up support for his wars.
Iran assassinating Trump would have some poetic justice, but I was against turning geriatric evil men into martyrs when Trump did it to the Ayatollah, and would likewise judge it net-negative when done by Iran.
Trump's actions did nothing to disabuse the Iranian regime of the notion that the US and Israel are their mortal enemies, but to be fair they already had that idea beforehand (and for good reason).
Still, I think that the regime is not so stupid as to see the US as a single entity. My guess is that they are a lot more anti-Trump than they are anti-US in general.
A future US president could offer Iran better terms than Trump can (because he must appear to have won his war), and Iran likewise has more incentive to offer them better terms.
I could be wrong, and it could be that killing their leader has put Iran on the warpath forever and as long as any IRGC member is left alive they will fire each missile at US allies as soon as they get it, but I doubt it. Iran knows it can not hope to wipe out the US, and so far has behaved somewhat rationally in foreign politics.
quiet_NaN
5d ago
Ok, I might have used Biden as a stand-in for DC more generally. (I do not think it was a signature Trump policy though, more like "in his first term there were still plenty of things running normally".)
I still think that the intelligence community had a rather good idea that if the US were to pull out, Kabul would fall to the Taliban rather sooner than later. Probably there is a heart-warming story of some small CIA analysist trying to warn the administration that the Afghan army will collapse immediately, and being confused at getting stonewalled.
quiet_NaN
5d ago
I think the Houthis will also want to avoid a scenario where Iran has to stop supporting foreign militias as part of a peace deal, which would only marginally better for them than Trump invading. And they would also not want to piss off Iran too much, because if Iran stops supporting then Saudi-bought bombs might eventually wear them down.
quiet_NaN
6d ago
Obviously, yes. For Iran, the victory condition in this war is "establish that attacking Iran is a presidency-ending mistake".
This was clear from the beginning, and is sound strategy on their part. The Trump administration is the main obstacle to a peace favorable to Iran. Trump started the war, so any concession to Iran will look like a defeat, so he needs to keep bombing them and hope that they will change their mind, and conduct an expensive escort operation to stabilize the oil price.
Any future administration will have the excuse "the Iran war was a stupid blunder". Unlike the occupation of Afghanistan, this is not something which the US will want to keep up for a few decades, even AIPAC will not be able to make them shrug off the high oil prices.
I think it is reasonable to presume that Biden knew that he was effectively signing over Afghanistan to the Taliban when he had the coalition troops leave, but was doing it anyway because he believed it was the right thing to do (which I concur on).
I am sure that a future president will find a similar fig leaf as Biden handing control to the new Afghan government. "Under this peace treaty, the US will leave the Strait of Hormuz and Iran will guarantee the freedom of navigation, and hand over some of their enriched uranium. Oh, now they are demanding tolls all of a sudden. Who would have thought! We will have to write a sternly worded letter to the UN complaining about that."
At the end of the day, if Iran manages to charge as much as $1.5 per barrel, that is a rounding error for gas station customers. And Iran would have some incentive to not take 10x as much because then it would risk getting bombed again (or annoying China, for that matter).
quiet_NaN
6d ago
I recently had a discussion with a heat pump owner and this seems to be a big issue. If you have floor heating and a wooden floor, then you very much do not want condensation, which would lead to your floor rotting.
For new houses, I think it is possible to design them with air conditioning in mind (so you have interior units which directly change the temperature of the air and pipe any condensation to the drain), but for a lot of preexisting homes things are more complicated.
quiet_NaN
7d ago
this is all just for show to build up hype around Anthropic, just like... basically everything they do
As someone who has used Claude, describing Anthropic as an entity which is only selling hype seems utterly implausible.
Are they building up hype and overpromising? Sure, just like any SV company.
Are they selling vaporware? Hell no.
For some workloads (e.g. "write a wrapper to stop that proprietary PoS program from segfaulting"), claude easily increases my productivity by 5-10x. Contrast this with gains from other interventions (larger monitor, good IDE, faster workstation) which are all much more marginal (typically much less than 2x).
The only thing which came even close wrt productivity increase for software development was "access to the web and a search engine".
quiet_NaN
7d ago
In terms of deaths per dollar spend, radiological weapons are likely terribly ineffective.
On the other hand, terrorists generally do not aim to win by killing all their enemies. Instead they aim to cause terror, which will trigger some change they think beneficial.
And for that, the fact that radiation is one of the primal fears of the modern human is kinda relevant. Anyone you kill through radiation counts a hundredfold compared to people killed by explosives or small arms. We are very good at detecting radioactivity, but not so good at judging its risk correctly. If after a radioisotope attack, the background radiation in some subway system is 5x of what it was before, people will not shrug it off. Instead, they will go full Chernobyl mode, and the cleanup alone will take vast resources.
quiet_NaN
7d ago
One time I did mushrooms. I had taken the same brand a time before and got very little effect. This time I did twice that dose and it put me in another reality. I eventually assumed I had dead and was in purgatory. Everyone around me I thought was demons. Suppose I got a knife and started stabbing my friends who I thought were demons. Should I go free?
In case where people deliberately placed themselves in states where they have little capability of telling right from wrong, I have little sympathy. Though by default, one might assume that any homicide committed is reckless manslaughter rather than pre-mediated.
But suppose that instead, someone had slipped the shrooms into your food without your knowledge. In that case, I would support sending whomever drugged you into prison for manslaughter and let you off the hook, unless the DA can conclusively prove that you were thinking "finally an opportunity to go on a rampage without doing time afterwards".
From there, the next step would be to consider some fictional insect which bites you and induces a homicidal rage. If the insect was so rare that nobody took any precautions against it, it would still be the same situation on your part -- you did not make any wrong moral choices, but simply were unlucky.
The practical problem with the insanity defense is that we can never disprove it completely. If some mafia hitman claims that he entered a psychotic state before any of his killings, that is not something which could be ruled out from forensic evidence. So pragmatically, we might shift the burden of proof a bit -- make it at least plausible why you did not know right from wrong instead of having the prosecution prove your internal state of mind.
quiet_NaN
7d ago
The current polity which enforces its laws over what is generally recognized to be Germany does not have your concept of "actual Germans". You can argue that naturalized citizens are not really citizens, just as you could argue that the federal republic of Germany does not really exist and you are instead the rightful Kaiser of Germany, but thankfully you would lack the power to enforce either idea (for now, at least).
Furthermore, I would argue that there is value in economic actors such as individuals, companies and states being able to make credible long-term commitments. If I sell you my house, I can't reneg on the sale three years later simply because I changed my mind. If an administration takes up debt, then there is an expectation that future administrations will continue to pay interest for it.
An exception might be if you argued that the entity making the commitment was not of sound mind when they made it. A five-year-old heir signing away their ancestral home for a box full of candy might be able to reneg on the sale later. If Angela Merkel had risen to power through a coup and granted the whole world citizenship in a fit of madness, that would be one thing. But in actuality, the governments which naturalized foreign residents arguably represented the will of the people. The option to not naturalize anyone was on the ballot, it was called NPD and consistently stayed below 5%. Until the AfD came along, having a naturalization path for non-Whites was a broad political consensus.
Not that the AfD would want to get rid of naturalizations either (just constrain it a lot more) and they certainly would not want to remove the citizenship of previously naturalized people (unless where the law already provides such a mechanism) -- at least that is what they say in writing, probably for the purpose of staying a legal party in Germany.
quiet_NaN
10d ago
Plenty of scope to still have high skilled immigrants (especially as a EU nation) without letting in massive family reunification et al.
Sure, but this is not what the median AfD voter wants.
Basically, if I were an Asian post-doc, I would stay the hell away from Magdeburg University (and most of East Germany, really). Sure, this could be overcome if incentives were high enough, but "the locals do not want you here, their state government (and sponsor of your employer) do not want you here" are big negative points for the tradeoff.
This would be weird, because open weight models which can do most of what the big three can do exist. I expect that SaaS is not exactly like finding exploits in existing codebases, where the proprietary models made good use of their few months of advance over the open weight models. If my startup model is to write smut on demand for my customers specific combinations of kinks and characters (which is probably not B2B, unless their kinks are really weird), then it does not really matter too much if I run on Mythos or Kimi K3. I would expect that most niche use cases are similar.
The big AI labs probably spend a lot of tokens identifying domains where their model's capability advantage is really decisive (like Navier-Stokes, or exploit-finding) and then a lot more tokens solving problems in these fields to showcase their model.
More options
Context Copy link