Whether 91% accuracy is acceptable depends on how unacceptable the 9% inaccuracy is. If 91% of the information in your term paper is correct you'll probably get a decent grade, but if you only kill 91% of cancer cells the surviving 9% will grow a treatment-resistant tumor and you'll probably die. This makes percentages essentially useless - more important is how badly wrong the worst wrong result is.
OTOH assumptions are cheap too, and I would venture to guess nobody in this thread knows anything about the hierarchy of the ADA, its most influencial insiders or their motives. We tend to shape those things according to the picture that satisfies us the most, and usually make it the simplest picture that invites the most hate. And saying this isn't an excuse or apologism, it's just staying open to improvement.
edit: And yeah, based on the douchevotes people seem to be taking my comment as support for the ADA, which it is not - they were definitely wrong to kick those guys out. Just saying all we know is what's in the article, and people who do the wrong thing can still eventually do the right thing if they're motivated.
Yeah you're not understanding what "hypothetical" means - it means suppose this happens, not asking if it can. Like if I asked how FTL space travel would change our culture, it's the culture part I'm interested in, not the physics being impossible.
This is fantastic if it means companies can switch to inexpensive ads that can be cobbled together by somebody's kids, and ditch their multi-million dollar ad budgets that just make everything cost more.
Have you tried OnShape? Web based, and the free account is full featured except you can't make designs private, which as a hobbyist is no drawback for me at all.
Brains. Anything to do with brains. Never had them but I once saw Graham Kerr, TV chef of the 70s and 80s, make sheep's brains on his show. I remember him saying they were very high in cholesterol. Of course we all know monkey's brains, though popular in Cantonese cuisine, are not often to be found in Washington DC, for what that's worth.
I think the interaction is a LOT better here - for one thing the community is so much smaller you get tons more replies here, which is more engaging and feels more being seen and heard. I still feel a strong one-answer-is-correct atmosphere, like you mention on .ml memes. But I guess that just means some subjects simply aren't discussable objectively, because that involves questioning your own ideas, which is seen as enemy infiltration. I'm slowly learning to avoid the echo chambers, but far from successful yet lol. Overall I heart Lemmy.
I know. I just think it's disingenuous to disapprove of fakery using AI but approve of fakery people do on their own. To me it's like saying you're evil if you use a ladder instead of asking a tall person to reach things for you.
Getting banned on reddit was how I first learned about lemmy, but tbh I see the same kind of shit-flinging here, and I go back and forth. There's a kind of smugness here about not being like reddit, but I think it just feels different because people who would be targets on reddit get to be flingers here.
Whether 91% accuracy is acceptable depends on how unacceptable the 9% inaccuracy is. If 91% of the information in your term paper is correct you'll probably get a decent grade, but if you only kill 91% of cancer cells the surviving 9% will grow a treatment-resistant tumor and you'll probably die. This makes percentages essentially useless - more important is how badly wrong the worst wrong result is.