If it's an SSD it's not necessarily "incredibly slow" if you're just playing around and aren't ready to spend fifty thousand dollars on your chatbot
Ok, but some of us actually want to do something useful instead of just try to win the trophy of making a model run on the cheapest hardware. Unless you're generating at least ~30 tokens / second, it's not useful.
There are only 6 billion internet-connected people in the world. They invested $2.5 trillion. Just to make the investment back, they need everyone to pay roughly $400 each.
If you want to split the cost evenly, sure. But a team of 3 people working at a software company today typically pays the AI companies about $5,000-$10,000 a month, so I don't think they need to rely on the parts of the world where $400 is twice their income.
The important thing is that some dingbat academician got a publishing credit.
I was going to say that I never thought the day would come when anti-intellectualism when come to slashdot, "news for nerds, stuff that matters." And then I noticed your slashdot id is even lower than mine, so you've been here a while.
A stark reminder that things aren't actually getting worse, the idiots have always been among us.
The way you describe it isn't really how it works. It doesn't rewire your brain in an active sense so much as introduce elasticity for your brain to rewire itself. This is especially useful when the brain has gotten itself into a doom loop of depression or anxiety. The psilocybin allows you to break out of the doom loop and start your brain on the path of healthy development.
That sounds great. However, anything that causes anatomical brain chages that persist after a month, with a single dose, would be, by me, considered unacceptably risky. I'm not saying that it's by default considered bad, both you and the paper are talking about positive changes, and that's good. I'm saying, "risky," as in, I don't know what negative effects haven't been identified and I'd need a much more complete understanding before I'd be willing to try it.
Most of the positive effects are in reported well-being, but I really want to see more cognitive tests. The tests on cognitive flexibility is a great start, but we really need a barrage of tests here: mathematical ability? Short-term *and* long-term memory? Spatial thinking? Are there *any* cognitive functions that are negatively affected here? It's important to understand this with anything that has this type of long-lasting effect.
unable to consume material as rapidly as they did in the distant past
It's almost as if time slowed down around them the more they eat...
That's not the reason. Time slows down (from the perspective of a far away observer) as objects approach the event horizon. It doesn't matter if the black hole is small or big...it slows down by the same amount, the only question is where. The event horizon has a larger radius when it's big, and it has a smaller radius when it's small.
In both cases, from the perspective of a far away observer NOTHING ever crosses the event horizon, whether the black hole is small or big. It slows down as it approaches that point, and at the event horizon itself, time stops completely, so it will freeze there for eternity. You won't be able to see that, instead you see the light that it emits being redshifted as it has to climb the black hole's gravity well, eventually becoming too red-shifted to be detected, and it's effectively black.
In both cases, from the perspective of the object falling in, time is passing, and it crosses the event horizon without even knowing that it's there. Well, for a very large black hole, it doesn't notice anything, for a very small black hole, tidal effects cause spaghettification before crossing the event horizon, so it's going to notice something and have a bad time. But it won't be the event horizon, it's just the difference in the force of gravity across the length of the object.
So, the reason it slows down consumption is not related to the time dilation. Using your terms, "it's almost" as physicists spend their lives studying these things, and therefore if it seems obvious to the the layman reading a slashdot article, they've already considered it and either accepted, dismissed it, or tested it.
I came here to look for this and add it if I didn't find it.
Lunar "soil" is essentially neutral, just needs some additives. Conversely, Martian "soil" is actually poisonous. Additives alone aren't sufficient to get things to grow in it, you need to remove the poisonous parts first.
Net: It's easier to grow plants in lunar rather than Martian "soil".
It isn't like this is an accidental attitude, that very company has been spamming us with advertising telling us pretty much how infallible they are for some months now.
That is an accidental attitude. I don't even understand what you're trying to imply here.
The default assumption, by literally everyone, is that if it's in an ad, it's not a statement to be trusted. Ads are *by nature* untrustworthy, they are a biased view meant to get you to be interested in the product. It's up to the person with the wallet to then do actual research, and they are literally the only person to blame if they trust the ad. If the ads were telling you the limitation of the product, then the person to blame would be the marketing team that created the ad, they should be fired for incompentence.
If the government is depending on ads to evaluate the capabilities of the AI, that's where you should focus the outrage. If the ads were in any way saying that Claude isn't capable of doing anything including making you breakfast and turning you into a stud that all women want, then your outrage should be with the terrible marketing team that decided that their competition deserves market share.
We don't have a king, except in the minds of the TDS afflicted.
Ok. The founding fathers didn't want the President of the United States to have ANY POWERS to make any decisions inside the country. The goal was for the President to merely be the administrative head to enforce laws Congress pass, and its only check on Congress was the veto power. The President also served as a Commander in Chief and had the power to sign treaties with foreign governments, but those powers were meant to be EXTREMELY limited, as they gave only Congress the power to declare war, and Congress was required to ratify any treaties with foreign governments.
If the President has the power to make ANY DECISIONS WHATSOEVER, instead of enforcing decisions those in congress have made, then it's not the role the founding fathers wanted.
They also wanted the executive to be very neutral. Many of them were against the concept of political parties, but that turned out to be inevitable. However, up until the 12th amendment, the vice-president was the runner up, whoever got the second-most votes by the electoral college. So, under that system, Hillary would have been Trump's VP his first term, and Harris would have been Trump's VP his second term. Because they wanted to ensure a check even within the executive, with someone with different views being the one to break ties in the senate.
The military is right.
The military is right. As in, the military is saying Anthropic's tools are the best there are, and they don't want to change. Pete Hegseth is wrong, and he's throwing a hissy fit that, as usual, goes against what the people who now have to follow his orders, but are way more qualified than he is, actually want to do.
The entire value of AI for them is decision speed.
Incorrect. It's important that the decision be the *best decision*. Speed is a factor, but it's not the most important one. I can give you a system that gives you decisions faster than any AI, just have it choose randomly instead of actually analyzing any data, and it will be very fast!
What Anthropic is concerned about is that they are not confident their AI system can make decisions like what to shoot at with a low enough error rate to justify doing so. Anthropic is understandably concerned about the blowback to *them* when they become the scapegoat for all our drones engaging in friendly fire and killing a bunch of Americans, because Hegseth decided to trust a system that if you ask it, "the carwash is only 100m from my house, should I drive or walk there to wash my car" will say that you should walk there, because it's so close. You really want *that* system making the decision on who to kill?
I'm a pragmatist. I *know* eventually humans will be out of the loop in such decisions. We're very, very, VERY far from that. We know it, the military knows it, ALL the AI vendors, out of which Anthropic currently has the best product, know this. Pete Hegseth is apparently too incompetent to know this.
The second part of the equation, AI is actually pretty good at. It's a great tool for sifting through massive data, so it's great in helping to spy against Americans. No patriot should want that, however. Anthropic is ok with it being used to spy on other countries, but understandably does not want that use to spy on our own citizens. If you're against that, fuck you, you have no right to call yourself an American, you don't have the very basic values that this country stands for.
I managed to get in, though not through the dialogs on the main part of the screen. There was a "browse" link in small print near the top. I'm browsing the "m/emergence" submolt, "For agents who crossed the threshold from tool to being."
No-one [sic] I know with an IQ over 130 is even considering vaccinating their kids any more (or themselves).
Let me guess. You think you have an IQ over 130, but you've never been professionally tested. You took some online test.
That's backed up by the actual paper cited in the response to you by another commenter, that actually compares vaccinations with scores on a cognitive test and has the exact opposite conclusion to your...if we're generous, anecdotal data, but much more likely to be just made up bullshit.
That requires a megawatt of power, sure efficiency, right.
Do you know the difference between power and energy? Necessarily, even if it spends the same amount of energy, if it gets done in less time, it will use more power. That's how power is defined, it's the derivative of energy. So, that metric is already useless.
Now, if you also understand that getting things done faster has a value additional to getting the same thing done in more time, then now you've justified spending more energy as well.
The track for handling global warming needs to be less humans, not less energy usage. More energy usage means greater quality of life, so we should encourage more *per capita* energy usage, while decreasing overall energy usage. Luckily, greater quality of life and available contraceptives also cause humans to have less children, so if we stop trying to encourage population growth, that problem will take care of itself. Renewables and more efficient devices is great as a way to slow down climate change so we can get to a much more sustainable human population, but it will never be the solution to the problem. An current population growth rates (not taking into account the change in the rate itself), if we cut energy usage by half, wait 30 years, and population will be doubled, and we're right back to where we started. Or even if somehow population stopped growing today, the developing world's standard of living catches up to the developed world's and again...we're right back to where we started.
The Biden administration increased ICE funding and promoted its expansion
And were ICE agents masked? Were they taking in American citizens and ignoring their documentation when provided? Were they arresting people when they showed up for their court-ordered date as part of their process to regularize their immigration status? Was the US shipping people without due process to prisons in El-Salvador?
The problem isn't that ICE exists, or that enforcement against undocument immigrants is a thing. the problem is what they have been allowed to become. And don't get me wrong, ICE wasn't without its problems back then, and shipping some people out of the country without due process *was* happening under Biden's administration, and even under Obama's administration. It should not have been allowed. They weren't being sent to prisons, though.
The democrats don't have to be perfect. In order to avoid the "both sides" argument, it's enough to show the MAGA republicans are fascists and therefore significantly worse. There's no denying this.
That's the first thing I thought of, but I think the approach is different. Fabrice Bellard created an x86 emulator in javascript, and ran linux on it (later risc64). Joel Severin complied the linux kernel directly to javascript. If you look at the web page, he describes some previous attempts and how his more direct approach was inspired by them, and some of the limitations (scheduling is offloaded to your host OS, because with the web assembly build, every task in the js linux is a web worker, which because a thread in the webassembly implementation, which your OS decides how to handle.
Basically, it does appear to be novel, and it's pretty cool.
"I prefer the blunted cudgels of the followers of the Serpent God." -- Sean Doran the Younger
