Jimmyeatsausage @ Jimmyeatsausage @lemmy.world

Posts

3

Comments

766

Joined

2 yr. ago

AI isn't giving the right misinformation

Same here. It's good for writing your basic unit tests, and the explain feature is useful getting for getting your head wrapped around complex syntax, especially as bad as searching for useful documentation has gotten on Google and ddg.

How science of everyone

For 50 years, science has been telling us that we're well and truly fucked if we don't do something about climate change. We didn't. I don't think the species is gonna die out (thought millions of other species will and are). You're talking like we're gonna be building a Dyson sphere (or swarm) in the next 20 years? Or colonize outer planets? It's science fiction...at least for next few 100 years. We can't rely on a deus ex machina save...we gotta take care of the planet we evolved to live on...if we aren't strong enough for that, I don't think we're getting anywhere near K-II.

Role a d20 and d4. If the d20 result is less than the caster's wisdom saving throw, collect

<d4 result>

Fuck you buddy, we're gonna cook ourselves first!

That'll teach ya...

When you train an LLM on a strictly curated and verified dataset that is limited in scope, it will do a very good job providing you with information about that specific topic and should hopefully give you the "As an LLM, I don't know about that..." speil for anything else.

When you let an LLM "do its own research" (e.g. train it on internet content) it starts telling you to put glue on pizza, eat a healthy number of rocks every day, and that you can run in the air as long as you don't look down.

Maybe they really are already as smart as people. /s

Everyone I know who's interested in raw milk probably has a few crates of ivermectin left over from the pandemic...should be plenty to keep them safe from the flu, too. /s

It's all a joke

If they want to live such a disgusting lifestyle, I guess it's their right, but they shouldn't shove it in everyone's face.

Queue Justice Thomas being seen, coincidentally, attending a large number of concerts with long-time friends on the Ticketmaster board...

Have you considered the possibility that you're a computer?

What would be better is polluting the software with invalid but still plausible constraints, so the chips would seem OK and might work for days or weeks but would fail in the field... especially if these chips are used in weapon systems or critical infrastructure.

It's a pretty big presumption that Elon Musk is providing transparent and accurate information to consumers about a technology he's hoping to sell. While I'd agree with the premise normally, he's kind of a known bad actor at this point. I'm a pretty firm believer in informed consent for this kinda stuff, I just don't see much reason to trust Musk is willing to fully inform someone of the limitations, constraints or risks involved in anything he has a personal stake in. If you aren't informed, you can't provide consent.

With some very particular opinions regarding hampsters.

It’s in the link I posted. They typed it up all nice with good words and a nice font and everything.

The article is several thousand words...none of which talk about a causal link or dose response in humans, which is the demographic I assumed the person I was replying to was curious about. It took me less time to find primary sources and link them than to read the biography of Ms. Hanson.

"I can't say for sure, but I'll know it when I see it"

I have, completely by accident and with no significant effort on my part, gone 40 years without mass murdering. It almost just happens on its own.

Propublica does good work on the whole.

Probably not yet...at least, not good, solid numbers. There's a study out of the University of Michigan that claimed to find that certain PFAS chemicals could double the risk for certain cancers in women with previous cancer diagnoses. Sounds from the abstract that it was just a correlational study (meaning it just shows a relationship between exposure and risk, but doesn't show that PFAS caused the increased risk...if you're interested in why a correlation doesn't establish causation, this site is a fun way to learn more https://www.tylervigen.com/spurious-correlations)

There's a lot of different PFAS chemicals and a lot of different cancers, so there's gonna be a lot of work required to nail it all down.

https://dceg.cancer.gov/research/what-we-study/pfas https://sph.umich.edu/news/2023posts/exposure-to-pfas-chemicals-doubles-the-odds-of-a-prior-cancer-diagnosis-in-women.html#:~:text=Exposure%20to%20PFAS%20chemicals%20doubles,prior%20cancer%20diagnosis%20in%20women