Sven Asai comments

No, it's not that bad, because it is important that the big accumulated money walks back from the big inactive accounts to the money circulation of economy. So a super rich person buying such an extremely over-priced item does something good also for everyone else. At least in UK in this example, because Sotheby's now pays more tax, gives more money to employees or secures their jobs, and they buy local foods or anything else, like a flower bouquet for their partner, and so on. For Japan it's not so good, that's right, because the Japanese winning bidder spent the money in UK. But who knows, maybe someone in that UK money circulation there also buys now something Japanese, a car, a Playstation or a package of Green Tea.

Spare all the daily unpleasant details, AI simply cannot work in general, because that is mathematically impossible right from the beginning.

The problem is complete negotiation incompetence. Normally a PM would play a little bit golf with Trump and most of all to fire such a miserable negotiator after 2 mistrials, not keeping that someone who couldn't reach anything in 8 attempts but even higher tariffs. What's been the problem with buying a little amount of US rice, meat, corn and a few Chrysler cars, and if not at all wanting to eat it or drive it then simply put it into silos and garages for emergency supply after floods or earthquakes etc. Trump happy, high tariffs avoided, everything fine for all other industries and GDP. But how can someone expect any useful thoughts or negotiation strategy from the current government? It's really a tragedy for the country if they stay in power. But it is now too late and the damage is done.

Well, it's Japan and so the Japanese people can install any legislation and penalty system they like or think it's helpful against crimes. Also the strong criticism is far from reality and logic. Why should they press innocent people into false confessions so that the real criminals still walk around and bring more and more dangers to society because not stopped? Doesn't make sense, right? Or the other complaints, detention centers and irrogation methods are too long and brutal and inhumane. Pardon? A lot of old and weak people like seniors for example, commit more and more low-level crimes like shoplifting and such, just for getting a safer place compared to normal outside life. No one would do that if those facilities were such an extraordinarily terrible horror as some here suggest.

calling for greater efforts to promote its application across industry and everyday life

There just can't be any serious usecases or applications in industry and everyday life, because it is mathematically impossible for this pseudo technology to work properly and with sufficient reliability. Yes, it's very interesting and nice to play with and I also understand that within this extremely hyped bubble many people now have found their jobs and income sources in this sector. But that all won't help to avoid all those 'AI' attempts to fail one way or another.

Saving quite some money in less useful sectors would be a beginning and could then be distributed to the affected people or for general tax decreasing. Do we need here space research? No, by biology we humans cannot travel or live in outer space. Do we need to send armed forces to Djibouti or navy vessels to visit harbors thousands of miles away? No, that is thrown away money. Do depopulated areas with high aged people a newly built bullet train access and stations? No, they are probably served better if a bus comes at least twice a day. And do after natural catastrophes devastated areas need a resettlement and luxury reconstruction? Probably not, because it may help with some nostalgia but again be destroyed in the next flood, earthquake or wildfire. The list surely goes on endlessly, but the point is just to start with some tax money savings and helping people in their daily struggle with higher rising costs.

I wonder why it is even necessary to cheat. This test is so extraordinarily easy, even I made it with 980 points.

It's fully understandable to begin the discussions early at a still moderate foreigner rate of 3%. As almost all other worldwide templates don't work , the only few exceptions are maybe Singapore, Monaco and similar rich or even more restrictive smaller countries, those generally too high numbers and randomness have lead to very unpleasant situations and crimes everywhere, for example bringing very far conflicts suddenly onto own soil when letting in both contrary folks, so it is necessary to make a framework that hopefully has a more positive and still bearable outcome.

Of course he's fully right. Although I'm a foreigner here and flew from the mass imported knife wielders, clan criminals and religious terrorists in Central Europe, I can fully support such opinions from native Japanese, even if they theoretically threaten my stay here in the future too.

Nothing special, and of course it will balance out very quickly. They now put massive tariffs on imported goods? The buying companies and customers pay the prices and then have less left for domestic produced things. They want bring in more foreign production capacities? Ok, then first, who will ever operate them, the shrinking old white men or the deported immigrants? Ridiculous, they don't even have the workforce for operation of foreign companies' based new or returning facilities, not in numbers and not in skills. Even if they had them, then second, who can buy all the many then made in US products? Not the own people, as they are busy and exhausted from overwork at all those plants and factories. Or do foreign countries now buy all the nice new US products? Also not, of course, as they just had seen their companies moving to US and leaving all left in poverty and mass unemployment. No, either it balances out such or it doesn't work at all right from the start.

That is the ancient times based cultural part of the show, but in fact has in most cases nearly not any significant influence on their chances or the election results. They could and should spare the useless effort and do their daily political work instead.

plans to delay construction of its second plant in Japan to prioritize investment in its U.S. operations to avoid tariffs

So the chips out of the planned second plant are not for the Japanese economy but for export to US? That's how I would read it. Well, then it's not a bigger loss. Of course a few jobs aren't created here , but I guess it's already difficult enough to have enough of the needed IC experts at hand and lure them into more shrinking and aging quasi rural areas where those plants are usually installed into, due to clean air requirements.

Environmental campaigners warn that deep-sea mining threatens marine ecosystems and will disrupt the sea floor.

That's ridiculous, compared to the available seawater covered surface, those few attempts and mini scratches are insignificant. The problems which cause massive usage of resources, energy and massive environmental damage are before and after that mining, like building ships and necessary machinery, transport of people to the ports and per ship into the mining area and back, then together with the minerals, and of course all the processes afterwards like chemical extraction, mass production of goods and such.

lack of demand

There is no lack of demand, there's only lack of money. They can immediately buy me a parking lot and a U.S. brand car at any price they like. So is the very easy deal, I bring what I have, my demand, and they bring, what they have, big money and the car. lol

If you’re using ChatGPT but getting mediocre results, don’t blame the chatbot. Instead, try sharpening up your prompts.

Of course, I blame it. If there's really any minimal AI in it, it would find itself a better or otherwise optimized prompt. Look, if we humans first have to find a good and usable prompt and afterwards have to check the results for useful content, reliability and truth, then the AI effect is completely neutralized and we can also directly search or create everything ourselves without need of an 'AI'.

was in no way connected to a manga author's disaster prediction

Ok, understood, so the manga predicted real one is still to come. rofl

Anyway, there is of course a certain significance but it is usually more symbolic, although the Upper House can be an obstacle, postponing some intentions of the government or stopping too big changes, and in theory it could even initiate a few own legislative attempts. It's maybe becoming a bit more interesting this time, if the ruling coalition really loses both chambers' majorities, but only then.

No one is immune against such predictions, because we are human beings and are interested in predictions and probable results of still unknown events. If it is a religious cult, a Nostradamus quatrain, a manga, a weather forecast, a money investment recommendation or an output of the newest AI model, we just want to believe or grab something good. And if it's not good like earthquakes in this case, it will still somehow satisfy us when we had predicted it right.

I have no clue how that should logically work. If after death the soul wanders into a child, then he already would know and could name the successor as he still maintains own soul. And communists determining the next head of a religion, that is very far from any logic either.

They must be out of minds. Electric shocks in a learning process? So what have those sick minds in store of us next, some leather belt whippings at the workplace for some higher productivity and beating the true potential out of everyone?

AI surely has an extremely big theoretical potential, but not any smallest chance in practice. Any serious mathematician knows it or should know it, as well as any serious expert in global social or society studies. I understand that the investments look more than promising and the theoretical capabilities are fascinating, but that's not how it works. There's no investment return from AI, because AI is mathematically impossible. It's been impossible in the 1940's and it's been impossible every time since when someone tried again in about every decade later, and it's impossible now and in future as well. But feel free to go ahead, everyone has another individual learning curve.

Yes, it's quite a potentially high price for keeping the affected politicians and NGOs in check. I guess they are the ones loudly crying here, not the real poor and vulnerable people in their poverty, famine and other extremely bad situations, because those have no voice heard, can't flee or migrate anywhere, don't have access to help at all, even if those organizations would have full or unlimited budgets.

There is no mathematical argument that says models have to fail

No arguments? Well, I have a few and a lot of other more famous people in the mathematical or IT sciences have them, the same or even additional ones. And there are even arguments from completely other fields, regarding society and social development etc. For example an aging and shrinking society cannot anymore handle the work amount of AI operating and checking all model outputs for plausibility etc.

So even if it according to you , would work very well technically, then it will fail anyway. We know and can or could show that it only can fail. That the other side doesn't believe and stubbornly tries with dollars in the eyes or becoming more and more angry investors in the back, that's not our cup of tea. Although we are affected too when a lot of money is wasted into something failing right from beginning, while it is needed for other global or society issues to solve. They all could know for decades from the first researchers up to contemporary living people in the field, that AI will fail.

That's of course a terror attack with life threatening chemicals and nothing else, not the deadliest and biggest possible, but still a terror attack. Now some may say, but the dose is only toxic for small lifeforms like insects. That can be true in most cases, but maybe not in all cases. The Sarin case also had victims who died and other victims who survived.

After all if the only factor was the input from humans that would be there in the first place and this would be observed in all models the same. This clearly refutes your explanation.

@ virusrex So, which other model input, different from given by human ones, do you have detected? From aliens? Self-generated (which would lead to falsifying of what we give the machines) ? Anything like that?

Anyway, I see it just from the mathematical side, and so I'm very sure that this whole attempt of AI development cannot other than fail , like it has failed before. It's all decades old (the underlying basics), and sometimes they gave it another try and put it back again, about every 10 years or so. And now they insist on walking the wrong proven paths a bit longer, but that's not my problem. They are all warned multiple times, that it cannot work, also not by scaling up and new data centers and their nuclear power stations placed every 500 meters on the whole planet, Moon and Mars too, or betting next on quantum computing etc. The massive investments are or will be lost, because no one can rewrite mathematical rules and laws, not they and not even we both. lol

Yes, a few degrees hotter, and I guess it's become difficult to deny. But I don't agree with what is made out of this empirical data in public discussions. We have constant or shrinking population, quasi de-industrialization, cleaner cars and industrial facilities, a lot of ecological activities in industry and private sector like less energy consumption, recycling or less packaging etc and still everything is put on us people with statements like we are the main and decisive factor for a massive climate change. That is not at all helpful and also can be very much doubted.

Sometimes neither of both is helpful. For example, in a nearby city they have a retro town, which of course should and wants to attract also more foreign visitors to bring some money in. But all directing signs are written in those romaji systems in the way the town citizens would say and write it, something like 'Returo matchi 500m', for example. I guess no-one capable of reading only latin characters or hiragana would be interested to go there, because it is not well to recognize that Returo matchi is very worth a visit as it is an interesting retro city area.

He owns the land and defends it from any intruders with a legally bought weapon, that would also be a possible and legal scenario.

It is very simple, natural behavior and biology say yes, against artificial constructs like states, legislation and law books which say strictly no. And here, if we like it or not, we can observe which one finally always will prevail one way or another.

"This is not just hallucinations. There's a very strategic kind of deception."

No, it's simply a reflection of ourselves, because the input data is from humans, or if you prefer from human 'intelligence', so the output is similar, regardless of being called artificial or generative. And if it is not one of the rare machine 'hallucinations', then it's just the stringently received decision by an algorithm, which is based on multiplying referenced numerical data by a structured or random matrix of weights. AI is just only a toy adults are playing with. One can compare it with a fine and sophisticated model car. Nice to have, to watch and to play with, but it's useless if you really want to use it as a car. It won't transport people or goods over long distances. The same here, AI is nice to have around and to experiment with it, but it never can or will work seriously and with acceptable reliability in any practical sense or scenario.