That threat to Putin is oversold though.
First, it's not immediately clear extra military aid will be a decisive war-ending move, an extra thousand armored vehicles or guided missiles will help, sure, Ukraine will certainly welcome it, but currently the greatest challenge for the Ukrainian military is a manpower shortage, that the extra metal does not solve. Even with "lend-lease on steroids," I don't believe it to be likely that Ukraine will be able to push the Russians back to the 2022 borders, maybe it'll just be enough to stabilize the front lines.
Second, Trump has an electoral mandate to reduce aid to Ukraine, not increase it. That's what he ran on. Putin knows this. This makes Trump's threat of increasing Ukraine aid a lot less credible. He knows that if Trump follows through with this threat, the American electorate will become displeased that the aid dollars doubled and yet the war still has no end in sight.
Well, I disagree. The primary driver for Democratic turnout in this election is Trump and abortion and none of them rely on Harris being anything other than not-Trump, you don’t need a deep-state conspiracy for her to be a viable candidate. Also I disagree that CIA brainwashed people to vote for her, like, I don’t even know where to begin with that.
Bernie lost his primaries fair and square. More people simply voted for Hillary, and then Biden. If he couldn’t even win a dem primary he’d been slaughtered in the general anyway.
I think this only applies if you believe the status quo will cause the society to invariably decline, I don’t believe that. And even if the US does decline in next four years under Kamala Harris, maybe the next president will be able to turn it around. In my head, it certainly beats the 1 in a million chance given to us by Trump.
It is an iterated game, maybe not with Trump, but with future candidates who might think to pull the same thing if the strategy works out for him.
N does not equal to 1. He’s done the same thing for his primary losses in 2016, and he refused to say he’d accept the loss in 2016 if he did lose. Also, whether n=1 is not important, unless you believe he’ll accept the loss gracefully this time around?
No, that’s precisely why it’s best that he doesn’t win. Trump is a defect bot for elections. He always plays defect, and everyone knows it. It's only rational that “it’s best for America if Trump wins” because the convention is to cooperate, and we know his opponent will cooperate. When defect-bots start winning elections because they only play defect, then all candidates will eventually become defect-bots and we all lose. So the most rational choice is to vote against the defector to ensure only cooperative candidates have a chance to win.
Well, from what I can read online, the Russians are paying upwards of 1.9m rubles (21k USD) for people signing up to become a soldier, and 5m (55k USD) rubles to their family if they die. It’s certainly more than the 30k number I gave earlier.
The thing is, right now you need 100 guys to drive the drones, but once they become autonomous, or semi-autonomous (imagine a drone leader and you just drive that one, and the others mimics/stays in formation autonomously), it’ll take a lot less manpower to achieve the same ends.
No, humans are not actually cheap and expendable, even if you approach this in a dispassionate way. Consider the economic costs for society for bringing up a soldier in the modern society. Food, education, child care, opportunity costs for parents, they all add up. How much is it? Online sources estimates around $300k USD in total in terms of monetary expenditure, but the societal costs probably a lot more due to opportunity costs for the “village” raising the child, but let’s go with the $300k figure for now. Not even accounting for government training, equipment, and logistics footprint, it’s already impressive. Say maybe a Russian, Ukrainian, or Chinese child takes less to raise, maybe they only require $30k USD in societal costs, that’s still a lot. If you can achieve a kill while expending a hundred $300 drones, it’s still worth it.
I mean, that battle had been fought and lost already. Software-as-a-service is the most prevalent thing now and no company will categorize this as you renting their software.
Currently the GenAI companies are not selling “paint” but offering painting services. You are telling them you want them to draw a picture of Mickey Mouse murdering Hillary in negligee with her feet showing, you are not drawing it yourself. The painter offering the painting service can certainly refuse requests that they feel to be problematic.
I don't see what I don't see :p
Yes, Chameleon is heavily lobotomized as a vision model, similar to SD3, but it doesn't mean a model like Llama-3 405B will be, given what we've already seen in the 8B and 70B versions. Of course, I could be wrong and the reason Meta hasn't released it yet is to first render it incapable, so we'll just have to see what happens when/if it gets released.
I also don't agree that the true objective of western LLM censorship is to suppress non-corporate AI capabilities, because the individual companies can just choose to not release powerful open-weight models at all, instead of releasing ones that are worse-than-useless. Simply because: if it sucks and doesn't move the needle, it'll be like it's never released because no one will use it, except the company also takes a hit to its prestige. I find it more likely that the objective of the censorship is indeed for the reasons they state, it's just covering their asses in case someone use them for objectionable purposes.
Ah I see. Your major concern about Western LLMs is censorship and not specifically power. Thank you for clarifying. I don’t think the 405B version will be too heavily aligned (e.g. closer to other Llama-3 models and not like Chameleon), at least not in a way that can’t be somewhat mitigated via prompt engineering or techniques like abliteration that someone already mentioned above. This is because it still has to be smart enough to hit the benchmarks, so they can’t afford to heavily lobotomize it.
Also thanks for expanding upon dense models vs MoE, that’s something I haven’t really considered.
Just a few hours after this excellent post, it’s announced that Meta is allegedly releasing the 405B Llama-3 on July 23 1, less than two weeks from now. Also allegedly it’s better than chatGPT on every single benchmark (hearsay from a protected Twitter account @futuristflower). The latter, assuming they mean GPT-4o, is less than likely, but still possible. If it turns out that it is actually more capable than any LLMs public or private available to us today, how does it impact your opinions, if at all?
It's a really good idea but I think this gets dangerously close to exposing how sausage is made to the general public. People will ask why the democratic party picked someone else over President Harris for the upcoming election, how would they respond to that?
I don't understand. If a scenario of murdering multiple candidates is asinine then surely the scenario of Biden murdering Trump is also asinine and the Supreme Court opinion doesn't matter?
If there's a complete break down of order that Biden is able to murder Trump without any major repercussions except some electoral issues in November then he can also do it to two or three other politicians that step up after Trump. I just took your scenario and expanded on it. If you have problems with the "if he's the only option left" part, that is not meant to be taken literally. As in, he's not literally murdering everyone who disagrees with him, just that no one of value will have time to establish themselves to challenge his presidency after the first few are gone.
Exhibit 0: Biden himself talked about his debate the next day. Are we supposed to be impressed about telling right from wrong? That he knows how to do the job he's been in for four years? These are not reasons to be elected President again, they are basic pre-requisites. For that matter, "speaking smoothly" and "walking" might actually be core requirements as well.
While that may be true. It’s clear he was contrasting himself with his opponent. He’s saying Trump doesn’t satisfy these basic requirements but he does.
Upon further research, you're likely right. This fact was not apparent to me reading through the reporting and discussions of this court opinion on other social media.
But is there a carve out for things "not necessary to the execution of his duty?" I don't believe there is, since that seems to directly go against this recent court decision. The judiciary wouldn't get to decide whether a specific implementation is "unnecessary" or "too-illegal." The president gets a presumption of immunity for all of his official duties, however he chooses to discharge them. The alternative just makes the decision moot because any prosecutor can just claim that a specific action is a "too-illegal" remedy to get an indictment.
So, yes, under this decision, I believe that, since the president can direct our troops, he can order Seal Team Six to execute his political opponents. The ordering part is the official act, and is immune to prosecution.
Ah you’re right. I was thinking along the lines of him admitting he’s done a crime on the broadcast, not that he literally commit the crime on the broadcast.
Does it even matter if he broadcasts his crime on prime-time TV? He could be there in his official capacity and thus the broadcast can’t be used as evidence for any wrongdoing.
Why stop there? Just order the guy after Trump whacked, then the next guy after that… Does it even matter if a large number of people refrain from voting Biden afterwards if he’s the only option left?
I agree with your points that this does not really have a material impact right now. But the prevention of authoritarianism also relies on a broader culture of accountability and respect for democratic norms in addition to the legal framework. I worry this opinion negatively influences this culture. I imagine the slide towards authoritarianism to be “death by a thousand paper cuts” so-to-speak, and this is one of the paper cuts.
- Prev
- Next
What question are we framing? Because if it's "Should transgender people be allowed to use gendered bathrooms of the opposite biological sex?" Then it's an unreasonable framing because trans people are not people with a cross dressing fetish.
More options
Context Copy link