I just consulted 54 trillion "people" who agree that this is idiotic.

[mook@possum.city]

@Natasha_Jay@tech.lgbt axios is basically an arm of the US state dept at this point, notice how they're getting some much exclusive coverage from the white house

[mnspy@mastodon.online]

@Natasha_Jay

WTAF

[knutson_brain@sfba.social]

@Natasha_Jay
“…no people were involved in the creation of these opinions…” ‍️

[lemonlolita@calckey.world]

@Natasha_Jay@tech.lgbt
If you ever wondered how polling could be less useful, reliable ...

[coolcalmcollected@mastodon.social]

@unlofl @Natasha_Jay

"how can it be worse?" is the same incantation as "what could possibly go wrong?"

and as we all know, incantations can only lead to tears

[bshn@hachyderm.io]

@Natasha_Jay it is unbelievable how some people can come up with such business ideas!
1. Buy one pricey GPU
2. Establish "polling company" entity
3. Collect orders, run simulations, earn bucks

[nothacking@infosec.exchange]

@david_chisnall @Natasha_Jay@tech.lgbt

... and LLMs are intentionally skewed to be "helpful" and not praise Hitler.

It turns out the average internet comment is rather nasty and people don't want that. The average internet user is probobly quite a bit better, but LLMs are trained on text, not people.

[doctormo@floss.social]

@knutson_brain @Natasha_Jay

That's got to be a book title in the future, right?

[ghostrunner@hachyderm.io]

@Natasha_Jay this DESPITE the foundational study's Conclusion that it is inaccurate.

https://openaccess-api.cms-conferences.org/articles/download/978-1-964867-73-1_45

[nilz@norden.social]

@Natasha_Jay

Is this SIMULACRON-3 ??

[knutson_brain@sfba.social]

@doctormo hmm…

[natasha_jay@tech.lgbt]

@ghostrunner
Why am I not surprised?

[slotos@toot.community]

@Natasha_Jay Are we automating Goodheart’s law now?

[slotos@toot.community]

@Robo105 @Natasha_Jay Correction, “what people that trained AI models think others should believe”.

[dukeboitans@mas.to]

@Natasha_Jay Fuck this shit. AI is garbage tech and completely useless. There's nothing to salvage here.
I say we nuke the entire site from orbit. It's the only way to be sure.

[priscillaharing@mastodon.green]

@Natasha_Jay run a simulation based on the other & former opinions of unspecified entities to determine the current and specific opinion of a certain group of people - yeah no

[peteriskrisjanis@toot.lv]

@Natasha_Jay this defies any purpose actual polling serves. Polling is quite literally with goal to get what people think at recent point in the past.
Polling already alone is incredibly hard, and internet and lack of land line phones has made even close to impossible. Anyone who has allowed such "llm generated polling" nonsenss to be published as truth should be called out to be liar.
Commercial polling has been struggling, but this takes the cake.

[overtondoors@infosec.exchange]

@Natasha_Jay Jesus fucking Christ. So... in other words, complete unadulterated bullshit posing as a survey?!?

[peteriskrisjanis@toot.lv]

@Natasha_Jay I understand llm appeal in content templating etc. but claiming that this somehow gives you any believable reference is madness.

[dtraslerwriting@mastodon.social]

@Natasha_Jay OK, this is why the other day I suggested that no one needs to quote "Elon Musk says" or any of that stuff anymore. Stop interviewing these idiots, and ask their idiot-boxes to make up what they MIGHT have said. That might get them to understand that having "AI" stand in for real people is a bad idea. And if it doesn't, at least we're still only getting bullshit answers like we would if we asked the "real" people.