Exactly this - and how chatGPT behaves too. After a few conversations with search enabled you figure this out, but they really ought to make the distinction clearer.
so your definition of "understand" is "able to develop the QC test (or explain tests already developed)"
I hate to break it to you, but the LLMs can already do all 3 tasks you outlined
It can be argued for all 3 actors in this example (the QC operator, the PhD chemist and the LLM) that they don't really "understand" anything and are iterating on pre-learned patterns in order to complete the tasks.
Even the ground-breaking chemist researcher developing a new test can be reduced to iterating on the memorized fundamentals of chemistry using a lot of compute (of the meat kind).
The mythical Understanding is just a form of "no true Scotsman"
Why stop there? Go all in: they should not run their open source totalitarian digital control nightmare codebase on closed source hardware, because that's the real issue!
A rational entity would want a functioning society, with working infrastructure and social systems, not a crumbling country on the verge of revolution. Guess what pays for that? Taxes.
Stop trying to rebrand selfishness as rationalism.
Do you voluntarily pay more than the minimum taxes required or are you selfish? And don't try to deflect the question by making spurious claims that wealthy individuals or corporations should be held to a higher standard. Give us a clear yes or no answer.
Look, I'm no economist, I don't know the exact numbers.
I can tell you however, that Bezos and his oligarch friends booking out a Venice that is sinking from climate change, while people suffer without health insurance and we still don't have high speed rail as built out as it needs to be is definitely not my understanding of fair.
And again, I say this as somebody that I also think should be taxed more. To live this well while so many people and projects need the resources more is unconscionable.
Leave the implementation details to the experts and elected officials. It's clear which direction we need to move in though.
So in other words you have no clue and are just making stuff up. If you want to donate more money to the government then go ahead, no one is stopping you.
That's not at all what I said and you know it. Shame on you for using such a bad-faith interpretation in your defense of the indefensible. The details of how we try to bend our capitalist system do not interest me because it is an inherently broken system. It is because of selfishness like yours that we will all die in the climate catastrophe.
Fair would be Bezos and his ilk getting the Jack Ma treatment, oligarch fortunes reclaimed by the people and used for solving the climate crisis. But we won't get fair. Meanwhile greedy little moneygrabbers argue over some little tax. Ridiculous.
Shame on you for making a bad faith comment you while refusing to be specific. It's horrifying that your would support anything done by communists. All communists should be killed.
I always used "prompting" to mean "providing context" in genral not necesarlly just clever instructions like people seem to be using the term.
And yes, I view clever instructions like "great grandma's last wish" still as just providing context.
>A given prompt may work poorly, whereas it would have worked quite well earlier.
The context is not the same! Of course the "prompt" (clever last sentence you just added to the context) is not going to work "the same". The model has a different context now.
well, there's your problem. it behaves like a search summary tool and not like a problem solver if you enable google search
reply