Simple: No other type of AI has "guardrails". I get that you are not smart enough for this level of deduction.

While the US is the uncontested master of climate-change denial, some of it is also going on in Europe. Stupid people enjoy a global distribution.

"Doubting" things is easy. And dumb fuck can do it. Verifying them or proving them wrong is hard. Especially when you do not even try ...

Just like hallucinations. No idea why people expect miracles from generative AI. It is not magic. At all. It is a small step forward, with some limited applications. Useful, but not "transformative".

Obviously, using a tool outside of what it can do well will usually do more damage than good.

What? That would be un-American! Clearly this "treatment" needs to be outlawed immediately!

I mean, a man, unmoving on the ground, not responding. Dead or unconscious. And they are still too cowardly to approach? How utterly pathetic and repulsive.

Funny how you just confirmed you have no understanding of things. Maybe, when no clue, STFU!

You are doing it wrong! You tried to do it and tried to get good results and then you analyzed what is happening. That is the wrong approach. Clearly, you should have just assumes LLMs are perfect and never verify whether they could actually perform....

That said, thanks for the description. I think what is missing to make the LLM perform is the tiny spark of insight that you can easily deliver via prompt, but that an LLM has no chance, ever, to generate on its own. The best it can do is have a matching input in its training data (or several that combine if they are frequent enough) and that is it. But generate it? No chance. The Math used cannot do it. LLMs are not AGI and that matters. Because it means they cannot even get tiny insights by themselves and they cannot fact-check at all because they cannot model reality competently.

On a meta-layer, you may have run into a guardrail they put in there so that the occasional code-base destroying misjudgment does not happen, because that would look exceptionally bad and destroy the illusion of competence. Hence they may be pushing "Human in the loop" here, even if that clearly does not work for many (most?) humans. It gives them somebody to blame if their product goes of the rails.

No solid engineering in sight anywhere. All these houses of cards will have to be torn down and replaced when it becomes obvious how fragile they are.

Always the same crap with the human race...

To be fair, there is a nearly endless supply of fools. I would not mind so much if these cretins were not making things worse for everybody else.

Completely untrue. You need to find out how a "central bank" works and what it does. Might stop you from making utterly stupid claims like this one.

To be fair, only the utterly clueless and the lying assholes ever made that claim. This thing is worse than useless, because all it does is create instability and burn value.

Probably. Their software is getting worse and worse and everybody hates being pushed into AI use. The last thing they need is a global security catastrophe from Win10 users.

Well, what did I expect. Obviously you would try to defend a deeply stupid stance with a fallacy.

There is a lot of applied stupidity in software design these days. The worst is when they try to force users to use features they do not want.