[sakoht]

> Convincing mimicry is good enough.

For what?

I ask questions and get wrong but “convincing” answers …that is way worse than wrong answers that are obviously wrong.

The functionality looks so smart because it skips reasoning and goes straight to “plausible imitation”. But the latter isn’t actually a path to the former.

If it were reasoning, and merely had errors the problem might-course correct. Even if the errors were huge, and reasoning was poor, it could eventually learn its way out.

It’s not intelligence. It’s a very good simulation of the superficial trapping of intelligence.

[beefield]

> For what?

Management consulting and Investment/stock analysis are the first ones that pop into my mind. (With some sarcsam, but admittably way too little)

[revolvingocelot]

>For what?

For replacing humans at any old arbitrary thing. It doesn't matter how actually wrong the output is, it doesn't matter that language models aren't actually intelligent, what matters is that it outputs 60% of human quality at 1% of the cost and .01% of the time (and that's at this early stage of language-model-development). That'll be more than enough to speed its adoption by nearly everybody, but certainly by bad actors, who aren't bothered by the "plausible imitation but not actually intelligent" aspects. "Clickbait SEO copywriting", for instance.

[janalsncm]

I asked my friend in what context a 90% success rate is considered “good enough”.

Pickup lines.

[LesZedCB]

watching an epistemological crisis play out in real time across the internet is truly sublime

[brookst]

Sure, those are all fair complaints about humans.

[koalala]

from the comments on the blog; couldn't this issue be solved by prime-prompting it with instructions to always lay out its steps before answering? this way the answer wouldn't be 'tainted' by the first few words being the wrong answer.

[hackernewds]

Same can be said about anything written by humans on the internet.