What statistical method do you base that claim on? The results presented match expectations given that Markov chains are still the basis of inference. What magic juice is added to "reasoning models" that allow them to break free of the inherent boundaries of the statistical methods they are based on?
Tobberone
My thought exactly. If consent isn't needed, what other actions do they deem justified without consent?
This is not a IP-issue, this is about human rights.
I agree. Trump just said that everyone that uses the services of American companies needs to agree with Trump or face the real risk of losing all their data.
In the face of that, even a pricier European option will look appetizing.
Not gonna lie, in the first half I thought you meant republicans.
Well, there is some data/rumours out there, stemming from a Dutch Tesla forum, that suggests that some fast charging might be beneficial for battery longevity. This seems to corroborate that. I can't remember the case for always fast charging, though.
That's an accurate name for the new toy, but not as fancy as "ai", i guess. Because we know that anything that comes out is gibberish made up to look like something intelligent.
You mean to say that Facebook isn't the main distraction of the -10s? Before Facebook it was TV that was the circus and low prices and no/hidden VAT provided the bread.
Well, I guess I do a bit of the same:) I do (70+2)(10+3) -> 700+210+20+6
Unfortunately i think it's the same old extortion we've seen before. Trump Jr already posted "I wouldn't want to be the last country to cut a deal", so I guess we will see more if this. Ukraine stood up against it, I hope the rest of the world does to, but I'm afraid most wount.
I don't know how we define "enough" in this scenario, but as you allude to: in the end the USA is just some 400 million people buying things from overseas. Absolutely those who buy the most, and that is what drives the economy in many countries. It is what has picked up countries in SE Asia from poverty to industrialization.
Problem is that now when everything is more expensive in the US, the same people will stop spending. They might have spent the same money on products made in America, but those are precious few and just increased in price. So in effect everyone in America can now buy less for the same money and the industry capacity to produce what's demanded doesn't exist in short term. And in real estate, short term is 3-10 years.
The rest of the world? Well, most of the world just lost their biggest market. Of course, the demand that can't be produced domestically in the US will still be seen, but at a reduced rate, which will reduce the economic development world wide, until new markers are found. China still needs to sell, but the market for the high margin stuff is reduced.
In the end? I wouldn't be surprised if this stunt reduces world trade to such a degree it might be viewed as a notable side effect that carbon use went down. Trump might have managed to stop overconsumption like nobody else and with it energy demand. So despite doing the oil industries bidding and go against renewables, the shipping industry stand to loose enough trade that it might affect oil use world wide.
Inflation is defined as the increase of prices over a set period of time. It is in itself nothing, doesn't do anything and its singular purpose is to be able to say how much something costs today compared to yesteryear. If the price difference depends on a supply chock (something that affects the ability to produce, like a shortage), or a demand chock (suddenly everbody rejects Tesla) is all the same, it results in a price change and can therefore be compared using the measure inflation.
Which method, then, is the inference built upon, if not the embeddings? And the question still stands, how does "AI" escape the inherent limits of statistical inference?