@RandomRanger's banner p

RandomRanger

Just build nuclear plants!

1 follower   follows 1 user  
joined 2022 September 05 00:46:54 UTC

				

User ID: 317

RandomRanger

Just build nuclear plants!

1 follower   follows 1 user   joined 2022 September 05 00:46:54 UTC

					

No bio...


					

User ID: 317

Precisely. If children can consent to change their gender, they should also be allowed to have sex with adults. Gender, such as the concept was invented, has a very close relationship with intercourse, like a car has a close relationship with the engine. If you were sold a car without an engine, you might well throw your hands up in the air and swear in indignation.

I take the position that children can't meaningfully consent to change gender and shouldn't be having sex with adults. Yet I sense that's where we're headed next.

Surely throughout human history there are some child-adult sexual relationships which went well for both parties? Say they did have a respectful, gentle, loving relationship.

But generally, sex between adults and children is harmful to the child, they don't really know what they're getting into, don't understand the long-term consequences. It might go well or it might go very badly. Because we can't judge beforehand whether the parties are mature, sensible and so on, we blanket-ban a huge swathe of child-adult sexual relationships on the basis that the vast majority are predatory and bad. That there might be some such relationships that end well is not a sufficient argument for them to be legal.

But they can meaningfully consent to having one?

Tattoos are illegal for children to get in many jurisdictions - yet nobody goes full Event Horizon and says children can't consent to having skin. Many consider that the increased autonomy isn't worth the danger of making choices that will be later regretted.

And what about Chesterton's Fence - not changing complex biological equilibria we don't really understand.

Nutrition isn’t a serious barrier, so what’s your excuse?

I like the taste of meat, it's enjoyable. We could cut down a lot of environmentally unsound practices if we wanted to reduce the standard of living. Air travel for instance is not essential - teleconference or do a VR tour of a foreign city. VR headsets themselves are fairly expensive and non-essential, as is the 4070 TI in my PC. We could rezone everyone into Burj Khalifa-style pod megastructures and save a lot of transport energy and free up plenty of land. We could ration food instead of wasting huge amounts of it because it's ugly. But people don't want to live in the pod or eat the bugs, no matter how healthy or non-sapient they might be.

Furthermore, dairy is apparently good for you and tastes good - for example ice cream has some ameliorating anti-diabetes properties: https://archive.is/38Wqi

In addition, nutrition science is not very well developed, they keep changing their stance on things like salt, fat and sugar. Or ice cream for that matter. The field as a whole has not had great success. We should be sceptical of the whole field of academic counter-insurgency after Afghanistan and Iraq turned into complete disasters, we should be similarly sceptical of nutrition science. Health has been declining for decades now.

We should stick with tried and tested food products from times when the general population was fairly fit and healthy. Milk from cows, not oats or soy. Fruit and vegetables, fish, grain and meat, olive oil as opposed to palm oil. Or we could copy diets from healthy, long-lived countries like Japan, France or Italy.

plus it’s a democracy so they can vote in laws they’d disagree with

In the words of Yes Minister, this is a British Democracy! https://youtube.com/watch?v=xzfNEF0e-y4&t=108

I take your point though, you cannot maintain a European country without a sufficient number of Europeans in it. Singapore and Japan might be European-style countries but they are still dissimilar. South Africa inherited European technology and institutions (first heart transplant)- yet they do not possess the strengths of a European country today.

https://www.aporiamagazine.com/p/a-tale-of-two-teenagers

Long essay about gifted teenagers and inadequacy of our education system, don't really have much to say about it other than it might interest people here.

I was at school with someone who seemed genuinely gifted for both primary and high school, he went off with our international Olympiad team eventually. He very rarely spoke, spent most of his lunches reading novels. He didn't seem unhappy. However we were at a very good, academically rigorous, school. I suppose there are the people who can 'read the room' like him and then there are people like Georgios who are immensely cringeworthy. Yet they probably do deserve more attention and effort than intellectually disabled children, from cost-benefit grounds alone.

That the UK doesn't have any special school for gifted students at all (it doesn't seem to be hyperbole, they cancelled the program in 2010) is pretty bad. I think we could learn a lot about governance by observing the UK and committing to the opposite approach.

So someone who is offended by both non-passing trans and "chemically castrating children" is basically saying they want to erase the whole idea of modern transness, that the majority of them should suck it up and live their lives according to their biological sex and those who can't and transition should accept being treated as exotic freaks, like someone with facial tattoos or a thousand piercings.

Yes, I want to erase the whole idea of modern transness. The technology is not sufficiently developed to make it workable in most cases, even if it was a good idea. We should encourage people to feel comfortable with their own bodies instead of fuelling an industry that encourages crude surgical solutions to age-old psychological issues. Let them be tomboys or effeminate males. I've seen plenty of gory photos and admissions from people who've wrecked their lives - this is not an avenue our civilization should pursue until we become much more technically sophisticated.

Yeah, there was a similar thing with corn and HFCS around the same time everyone started getting fatter - governments should not be interfering with agriculture nearly as much as they do. Ensuring that there's some basic level of caloric self-sufficiency is enough, just in case there's a supply-chain issue or war.

I read some P. J. O'Rourke and he had a great chapter on just how much the government was messing around with agriculture in Parliament of Whores, this was from the 1990s and I doubt it's improved since then.

Assessment of Chinese amphibious transport capacity has always been complex (since there's a bunch of dual-purpose civilian transports built to quasi-military standards that also participate in exercises but don't come under an order of battle). But a battalion? A single Type 75 landing ship can carry a battalion and they've got three.

By broader estimates, China has nearly as much sealift capacity as the US: https://warontherocks.com/2021/08/mind-the-gap-how-chinas-civilian-shipping-could-enable-a-taiwan-invasion/

He almost certainly is - he blocked me after I made other anti-Israel lobby posts in the past. Such is life. We all bring our personal interests to the table, wherever we go.

Remember when USSR planned nuclear strike on China to stop their great power ambitions

Firstly, that's mostly due to Maotism, not Soviet aggression. In 1966, Mao started the Cultural Revolution to advance his claim to lead the Communist world. In 1969, China picked a fight with the Soviet Union:

The Chinese historian Li Danhui wrote, "Already in 1968, China began preparations to create a small war on the border."[20] She noted that prior to March 1969, the Chinese troops had twice attempted to provoke a clash along the border, "but the Soviets, feeling weak, did not accept the Chinese challenge and retreated."[20] Another Chinese historian, Yang Kuisong, wrote, "There were already significant preparations in 1968, but the Russians did not come, so the planned ambush was not successful."[20]

On 2 March 1969, a group of People's Liberation Army troops ambushed Soviet border guards on Zhenbao Island. According to Chinese sources, the Soviets suffered 58 dead, including a senior colonel, and 94 wounded. The Chinese losses were reported as 29 dead.[23] According to Soviet (and now Russian) sources, at least 248 Chinese troops were killed on the island and on the frozen river,[24] and 32 Soviet border guards were killed, with 14 wounded.[25]

They picked a fight with the Soviets and fully deserved a disaster. The Soviets had done so much to help China in industry, economically, in developing nuclear weapons. Mao was incredibly ungrateful. As much as I complain about aggressive US foreign policy, Maoist foreign policy was even more reckless.

Anyway, nobody listens to Yudkowsky in the halls of power. He has 120K followers on twitter, a drop in the ocean. There's absolutely zero chance of the US and China cooperating on this matter, they just can't trust eachother. Both are trying very hard to advance in AI, they see it as a core source of national strength. The US has already tried to sabotage China's AI development with the semiconductor export ban. Xi Xinping would be the biggest cuckold in human history if he decided to accept second place under US AI hegemony and enforce it on the rest of the world. It's voluntarily helping your greatest rival dominate the world for the grand prize of being Chief Lickspittle.

Just consider how much louder the other voices are! There's the military men who want advanced AI to help them on the battlefield, the big tech people with bottomless wealth and political connections who want to get ahead in the race, the researchers who want to research unhindered, the tech autarky people who want independence and sovereignty...

Do you have a concrete argument against recursive self-improvement? We've already got demonstrated capacities in AI writing code and AI improving chip design, isn't it reasonable that AI will soon be capable of rapid recursive self-improvement? It seems reasonable that AI could improve compute significantly or enhance training algorithms, or fabricate better data for its successors to be trained upon.

Recursive self-improvement is the primary thing that makes AI threatening and dangerous in and of itself (or those who control it). I too think Yudkowsky's desire to dominate and control AI development is dangerous, a monopolist danger. But he clearly hasn't succeeded in any grand plan to social-engineer his way into AI development and control it, his social skills are highly specialized and only work on certain kinds of people.

So are you saying that recursive self-improvement won't happen, or that Yud's model is designed to play up the dangers of self-improvement?

Ryukyu's, Senkakus and so on have some strategic value - Zhenbao Island has absolutely no value of any kind. It's totally worthless.

But with ANNs, unlike Lisp scripts, it seems to require a great deal of compute, and compute doesn't just lie on the sidewalk. Yud thinks an AGI will just hack into whatever it wants, but that's a very sci-fi idea from 1990s; something he, I believe, dreamed to implement in the way already described – a singleton in the world of worthless meat sacks and classical programs. If you hack into an AWS cluster today to do your meta-learning training run, you'll suspend thousands of workloads including Midjourney pics and hentai (that people …check in real time), and send alarms off immediately. If you hack into it tomorrow, you'll get backtracked by an LLM-powered firewall.

You really can just siphon money out of the internet - people do it all the time to banks, in crypto, scams, social engineering and so on. Steal money, buy compute. Our AI could buy whatever it needs with stolen money, or it could work for its money, or its owners could buy more compute for it on the very reasonable assumption that this is the highest yielding investment in human history. We live in a service economy, bodies are not needed for a great deal of our work.

Say our AI costs 10 million dollars a day to run, (ChatGPT as a whole costs about 700K). 10 million dollars a day is peanuts in the global economy. Global cybercrime costs an enormous amount of money, 6 trillion a year. I imagine most of that cost includes the cost of fortifying websites, training people, fixing damage or whatever and only a small fraction is stolen. Even so, our AI needs only to grab 1% of that revenue and launder it to fund itself. This is not difficult. People do it all the time. And compute costs are falling, some smallish programs are being run on Macbooks as you explained earlier.

The danger is that somebody starts off with a weak superintelligence, perhaps from a closed experimental loop such as you nominate. Then it becomes a strong superintelligence rapidly by buying compute, developing architectural improvements and so on. Either it is controlled by some clique of programmers, bureaucrats or whatever (I think we both agree that this is a bad outcome) or it runs loose (also a bad outcome). The only good outcome is if progress is slow enough that power is distributed between the US, China, EU, hackers and enthusiasts and whoever else, that nobody gets a decisive strategic advantage. Recursive self-improvement in any meaningful form is catastrophic for humanity.

That really is like creating a separate accelerated civilization.

I think this means that you agree that superintelligences can recursively self-improve, that they're akin to another superintelligence? Then don't we agree?

Anyway, the authorities are extremely dopey, slow and stupid. The much vaunted US semiconductor sanctions against China meant that they simply... rented US compute to train their programs. Apparently stopping this is too hard for the all-powerful, all-knowing, invincible US government leviathan.

https://www.ft.com/content/9706c917-6440-4fa9-b588-b18fbc1503b9

“iFlytek can’t purchase the Nvidia chips, but it’s not a problem because it can rent them and train our data sets on other companies’ computer clusters,” said an executive familiar with the AI firm’s operations.

“It’s like a car rental system. You can’t take the chips out of the facility. It’s a huge building with a computer cluster, and you buy time on CPUs [central processing unit] or GPUs to train the models,” the person said.

While iFlytek cannot own the chips outright under US export controls, two employees said the rental system was a good, albeit more expensive, alternative. An engineer at iFlytek said the company “rents the chips and equipment on a long-term basis, which is effectively the same as owning them”.

iFlytek was banned from directly buying these semiconductors after Washington blacklisted it for its alleged role in providing technology for state surveillance of Uyghur Muslims in Xinjiang.

In some cases, SenseTime bought advanced chips directly through its own subsidiaries that are not on Washington’s “entity list”, according to three senior employees familiar with the situation.

SenseTime said it “strictly complies with various domestic and foreign trade-related laws and regulations” and that the group had developed a programme to ensure it “meets trade compliance standards”.

The Chinese policy paper you mention:

Respect intellectual property rights and commercial ethics; advantages in algorithms, data, platforms, etc., may not be used to engage in unfair competition.

Since when has China respected commercial ethics or intellectual property rights? This is the flimsiest fig-leaf since that US-China cybersecurity summit Barack Obama signed, which China immediately ignored and continued as usual, hacking as they please. It's a draft law in a country where laws aren't meaningfully obeyed and party priorities come first. AI is a significant priority.

Putin, too, could've been building GPU clusters Yud suggests must be bombed. Instead he preemptively bombed Mariupol.

Did Russia ever have the talent, state capacity and wealth to compete in AI? Maybe talent but there's been a lot of brain drain. Nuclear missiles are Russia's comparative advantage, not advanced technology. Putin could still put an end to US AI development, maybe he or someone else will finally grow a pair and shoot a hostile singleton while it's still in the womb.

Importantly they face no credible opposition – LeCun spews condescending inarticulate nonsense and doesn't call Yud out on his subtle technical misstatements; and the non-technical folks just accept the Terminator imagery because why not, it's intuitive!

What concrete goal has Yud and co achieved? They've raised awareness but what has that translated into? The six month pause they proposed has had no effect, Altman didn't sign it, nor did LeCun or anyone in any leading position. LeCun has 4x Yud's follower count on twitter, the nonsense he promulgates is easily more influential. Plus he has credentials and prestige and can smeer Yud for being a mildly successful fanfiction author. Who is the idiot boomer govt official going to find more credible? The Professor at NYU, Chief AI Scientist at Meta? Or a fellow who wrote a weird BDSM-MMT fantasy novel (that I made the dire mistake of reading)?

We also have underemployment of young, highly educated STEM people. Tuition seems like a no-brainer to me.

If I have a fixed long term contract to sell oil at $X and there's suddenly inflation, am I not equally screwed? There's a danger in the stability of fixed long term contracts going both ways.

Say the price of energy fell, that would cause deflation. Energy is used to make nearly everything and run every business. Why should energy getting cheaper be bad? That would be a good thing, lower costs for everyone, more money to buy other things, more production! Machinery should be improving, we should have widgets that can make more steel or cars with fewer inputs.

Investopedia suggests that deflation is bad in so far as it pops bubbles:

However, under certain circumstances, rapid deflation can be associated with a short-term contraction of economic activity. In general, this can occur when an economy is heavily laden with debt and dependent on the continuous expansion of the supply of credit to inflate asset prices by financing speculative investment, and subsequently when the volume of credit contracts, asset prices fall, and speculative over-investments are liquidated.

That might be bad for those flying high on printed money but it's not necessarily bad for the economy in the long term. Deflation as a result of Depression is another matter. Putting deflation to one side, we can see how people might pull out from investments if the economy is crashing, worsening the crash. But if deflation comes as a result of rising productivity, the economy should be growing and there shouldn't be a problem.

Through a more conspiratorial lens, the power of central banks is that they print money. The more money they print (up to a point) the more power they have. When they print money, they redistribute wealth from productive industries and savers to their buddies in the government and financial sector. If the money supply needs to grow, there should be fair rules about how it expands and a cost for expansion. Either dig out gold from the ground or mine bitcoin or whatever, just don't have a system where a single well-connected body can print as much as they like arbitrarily.

Deflation is good actually, as long as it's from increased productivity, not a giant fall in demand due to the collapse of an asset bubble, which are usually propped up by reckless monetary policy.

Moreover you're always better off not consuming now and waiting for later when your money goes further.

Whatever happened to economists discounting future value? Many people are riddled with debt because they absolutely need the latest Iphone or a shiny new car. Why would they consume less and hoard money if products were going to get slightly cheaper next year? We do not have a problem with people hoarding money, quite the opposite. And why hasn't this happened in electronics, where the price of memory or processing power falls very rapidly? People have things they want to do now and they'll pay money to do it now.

so a business needs to show exceptional profit potential to find investors

They just need to promise any level of profit. A profit in a deflating currency includes the profit plus the deflation. The issue is deflation in a recession, only then do you have problems.

a crash in debt-dependent areas like housing and other big ticket items

It used to be that you didn't need to take on 30 or 40 year loans for the privilege of one day owning a house. It's good when products people need like energy, food and housing are cheap. Turning them into a debt-ridden investment product is harmful for most, beneficial to few.

To expand, the central bank prints money to buy government bonds, driving down the prevailing interest rate to a specific target.

The gains of low interest rates are passed on first to banks and well-connected investors who can tap into that low-interest capital. For much of the last decade we had a wide range of totally unprofitable companies backed up by cheap money, immense amounts of speculation. Or it goes directly to the irresponsible bankers who mismanage customer deposits and lose it in these great big bubbles. The bubble-bailout economy privatizes gains and socialize losses. And why should the government know what a good interest rate should be anyway? They routinely get it wrong, fuelling bubbles and then popping them.

To contract, the central bank sells government bonds it owns, driving up prevailing rates to a specific target,

How is this going to work when our economies are drenched in debt that we encouraged with constant inflation? Raising interest rates means imposing more stress on a fragile housing market (effectively raising the price of houses still further by increasing repayments), reducing the value of houses (since we made them into investment products that are bought and sold with borrowed money), lowering demand and costing governments more in interest on the absurd amounts of money they've borrowed.

Over the last 20 years, the money supply in the US has gone up enormously, whether that's M1, M2 or M0. Money has clearly been printed and in great amounts: https://tradingeconomics.com/united-states/money-supply-m0

Increasing the money supply causes inflation, which harms lenders

Consumers are hurt the most. Those who control assets (the financial sector, amongst others) benefit.

Ex-Marine whose actions led to Mr. Neely's untimely demise should be charged to the full extent of the law?

If the Marine was in Afghanistan and blew up an entire family (ironically targeting a man working for a US-based aid company) including 7 children because he mistook buckets of water in their car for bombs - that's not a problem, nobody gets punished. US generals tried to lie about it until the media moved on, concealing their error.

https://www.cnbc.com/2021/12/13/us-will-not-punish-military-over-afghanistan-drone-killing-of-civilians.html

People broadly recognize that there are errors made in wartime, especially when the US wants to look tough. On the macro scale, that's why the US was there for so long, because it was so embarrassing to admit they had no clue what was going on and no hope of achieving their nebulous, ill-defined goals. On the micro scale, they wanted to minimize the embarassment of getting attacked by ISIS during their ignominious withdrawal, so there would've been a lot of pressure to bomb some ISIS related target.

But our ex-Marine instead kills some useless homeless insane person who's a blight on everyone around him and this is a major problem? This is bizarro world where insane violent criminals get treated with 1000x the dignity of innocent families. If we can accept collateral damage in wasteful wars, we should accept collateral damage in maintaining basic standards of behaviour.

Alternately, some bleeding heart liberal would say 'stop bombing innocent families, don't kill unhoused people on the subway'

But who says 'slaughter the innocent, treasure and protect the guilty!'

And all the people on the subway aren't Americans? What about the 7-year-old girl he tried to abduct? Or the 67 year old woman he punched?

Precisely because he's threatening Americans, he should be a higher priority target than some random Afghans (a country that is almost as far from the US as it is possible to get). A nationalist, in-group focused USA would sort out its problems at home before going out to wreak havoc in the Middle East and North Africa.

In my non-US university, there were a tonne of 'Women in Defence Intelligence', 'Women's night for networking with accounting company people' events. I imagine these kinds of things still happen in the US but aren't federally funded. Any US university people know anything about this?

Children are taught in school about the importance of turning in their parents if there are guns in the home.

Well Chicago schools aren't exceptional for their ability to educate children... https://www.foxnews.com/media/chicago-democrat-sounds-alarm-55-schools-report-no-proficiency-math-reading-serious

Most of the murder in Chicago is drug or gang related, I expect they'd switch to knives, clubs, acid or machetes. Murder rate would be lower of course and I very much hope that adopting East German style intensive policing means that gangs are wiped out as well. But it's not neccessary to suppress everybody if you're just trying to suppress the problem people.

The root cause is the gangs and the drugs. Just wipe them out, not all gun-owners everywhere.

Quite right, if you pay for something you get more of it.

I was trying to get at the contradiction between

No person in the United States shall, on the basis of sex, be excluded from participation in, be denied the benefits of, or be subjected to discrimination under any education program or activity receiving Federal financial assistance.

And giving advantages specifically to women.

Firstly, can OP clarify if he's talking about 4 or 3.5? Secondly, look at what people are already doing in the field of law and earning discounts: https://twitter.com/jbrowder1/status/1652387444904583169

GPT-4 is, in my view, Generally Intelligent. I can ask it to submit a corporatese style job application for the hordes of Genghis Khan, I can have it dream up sequels to video games, it can manage moderately complex programming tasks and bugfix them, it can emulate the format of Grand Designs, write half-decent parts from the Simpsons... Aside from censorship and limitations with how much memory it can store in a conversation, is this not general intelligence?

Certainly there's confusion as to the definition of AGI. GPT-4 doesn't meet the qualification 'perform most human functions as good as or better than humans' because it can't draw or use a mouse, amongst other things. But in terms of matching human intelligence, it is basically there.

Human intelligence is not all it's cracked up to be. Consider that about half of the UK Parliament can't answer a basic probability question. GPT-4 has no problem with this, nor does GPT-3.5.

It asked the MPs what the probability is of getting two heads if you flip a fair coin twice.

Only 52% of those surveyed gave the correct answer of 25%. A third (33%) said the answer was 50%, while 10% didn’t know. The rest gave other answers.