I’m probably the most anti-AI person I know, but I agree discourse around how “AI is theft” is a bit shallow.

Copyright is often erroneously conflated with plagiarism. While the two do sometimes coincide, they’re very different concerns.

I, myself, believe copyright is so broken we’d be better off throwing it away. (The only thing I believe I’d miss about copyright if I woke up tomorrow and it didn’t exist would be copyleft.) But I do deeply believe in a right to attribution. I don’t think AI is theft. I think it’s plagiarism.

And I believe that listing the names of all those whose works were included in training data for a model would still be a great disservice to the artists buried tens of millions of names deep right after some dumbass “NFT artist”. Meanwhile, asking an LLM or image generating model which training data was involved in generating one particular piece of output it produced is futile the same way as asking a stage strongman which rep at the gym allowed them to lift that car.

And if someone objected that giving what I would consider “sufficient credit” to artists/authors/whoever would make AI models completely infeasible, then my response would be “that’s exactly my point.” If it can’t exist without taking advantage of huge numbers of people without their consent, then it shouldn’t exist at all.

Finally, one more point I want to make is that if AI didn’t make billionaires a huge amount of money, the legal system would have put a stop to the mass scraping of training data and made a very visible example of whoever undertook to do mass scraping in the first long ago. (Never forget what they did to Aaron Swartz for scraping on a vastly smaller scale than OpenAI or Twitter or whoever did to make their LLM models.) As terrible as it is having to deal with the shitty IP laws we have, the greater injustice is that the laws (IP and otherwise) only apply when billionaires want them to.

“By not drinking antifreeze, you’ve just re-sensitized yourself so any antifreeze you drink will be impossible to metabolize.”

Is there any particular piece of information that he revealed which could have been used by anyone really to… I dunno… bypass defenses or take advantage of people or whatever in some a way that could actually hurt people?

I dunno. Everything I’ve heard is that everything that he leaked that has been released was super innocuous militarily (not that the military is a bunch of knights in shining armor or anything) or national-defense-wise. It is (or at least should be) very embarrassing to the U.S. “intelligence apparatus”. And it’s clearly good reason to believe that Uncle Sam clearly doesn’t have our (American’s) best interests at heart. But what could possibly have even hypothetically been used to cause any harm?

(And, I don’t know, maybe you know something I’m unaware of, but it really seemed like he went out of his way to avoid any harm to anything but the reputation of the intelligence industrial complex. And maybe a few presidents.)

made us more vulnerable to threats

How so?

How so? Like, is the content super gritty in South Korea? Is the EULA super fucked-up there? Does it cost an arm and a leg? Is it unavailable without a VPN? Something else entirely? All of the above?

Standup comedy is all about personality and “vibes”. What they’re drinking affects the audience’s perception of both. It helps to set expectations. Their drink is often part of the performance. (Mind you, comics don’t always have beverages. I watched a comedy routine today and I’m pretty sure he didn’t have a beverage at all. Not even water.)

Reddit owns/uses multiple domains and the multiple domains can certainly collaborate to compromise your anonymity (or at least pseudonymity) even with third-party cookie blocking enabled. For instance they use redd.it for url shortening like this url. You could clear all your cookies for *.reddit.com and perhaps hide that you are the same user they previously banned for a while, but if redd.it has ever set any cookies that you haven’t cleared, the first time you visit a redd.it it could reveal your identity to Reddit. They probably own quite a number of other domains that might similarly reveal your identity to Reddit, and there are most likely third party companies that own other domains that will collaborate with Reddit in some way or another that will reveal your identity to Reddit.

As someone who has been boycotting Reddit since the API-enshittificationing, I can’t help but echo SpaceNoodle.

However.

Did you make sure your IP address has changed? Typically dynamic IPs don’t change very often unless you disconnect your router and/or modem temporarily (and sometimes even that won’t do it.) You’d need to check what your public IP is, restart or disconnect-and-reconnect your modem to get a new IP address, then check your IP address again. Unless you do all that and confirm your IP address has changed, it’s risky to try again on your laptop.

Also, you’ll want to clear more than just your Reddit cache. To be safe, I’d recommend clearing all your data in your browser. Cache, cookies, history, local SQL, all of it. And for all domains. (I don’t know to what extent Reddit may use other domains for things like authentication and such. They might still be able to tell you’re “you” even if you clear all the cookies and such for *.reddit.com.)

Meta’s months-old AI unit is a soul-crushing gulag

But they repeat themselves.

More like “nice-in-quotation-marks”. Maybe “polite” would have been a better word. They smile bigger – exaggerated and insincere. They try to satisfy whatever you’re wanting so they can conclude the interaction as quickly as possible. They keep an air of superior self-righteousness while being nice… defensively.

Mind you, I do even put “polite” in quotes on purpose. The more stuck-up folks can very “politely” cut you straight off at the knees in a way that preserves some semblance of plausible deniability where poorer folks (on average, very much a generalization) may be more real/authentic in general and can be more direct about calling you an asshole.

Mind you, this friend of mine who made this observation didn’t say any of what I said in this comment. That’s just my own editorializing.

No matter how low it goes I’ll never understand how it isn’t zero. (As much as it should be zero, it unfortunately never will be. Well over a hundreds of years from now, I imagine right-wingers will romanticize Trump the same way they romanticize Reagan and the Civil War and… you know… Hitler.)

I have a friend who recently moved from the poorer side of the town I live in to the richer. (I’ve always lived in the latter, richer area.) And he was talking about how striking it was to him that on the richer side, strangers would go way out of their way to avoid interacting or making eye contact or anything, but the moment you forced the issue and start actually talking to them, they’re the nicest people ever. The other side of town, interaction wasn’t avoided the same way, and they wouldn’t necessarily be quite as nice in their interactions.

Wow. That’s super shitty that’s becoming necessary for some projects. But I don’t fault the Ladybird folks for deciding this was necessary.

Making it possible to vote up and/or down while preventing double-voting for the same account in a distributed application like Fediverse sort of applications seems challenging at best unless every instance knows exactly who voted which way on what. What I wish is that Lemmy would make it a proper feature that it was public information who voted on what.

I dunno. Maybe it would be possible to implement some zero-knowledge proof sort of thing that would keep people from double-voting (purposefully or accidentally) without anybody but the voter knowing who voted for what. But absent that, I’d rather that seeing that information wasn’t limited to an elite group of users composed of just mods/admins. I’d also rather that I didn’t have to go to a separate site to see information about upvotes and downvotes.

I suppose the argument could be made that we could get away with not having votes. Just make how high it shows up high on the “hot” sort by how many comments it has. Though I do feel like there are “good” posts that I’d want to see with few/no comments.

Ha! I don’t mod many communities, but I like to think I’m pretty reasonable about it. I definitely try to err on the side of not taking mod action.

I have a very selective list of “I want to see all the posts that get posted to this community” communities that I’m subscribed to that gives me like 5 posts per day at most, and I sort that by subscribed/new. Once I’m done catching up with those, I go to all/new, and I block communities I don’t ever want to see again. It’s nice because 1) I get to see all the posts I really want to see, 2) I can block the “crap I don’t want in front of me for no reason”, and 3) new communities that I might be interested in make it in front of my eyes without me having to go search for them or hope the ones I’d want to see happen to come up in a “new communities” community on my feed.

Thanks for this! I’m not sure it’s working correctly for me. It’s only showing five downvoted posts. (And according to lemvotes.org, I’ve downvoted over 300 posts.) After that it gives me an error: TypeError: can't access property "data", W.value[r.feedId] is undefined and the “retry” button doesn’t do anything. But it still gives me something to try if I can figure a way to fix that somehow.

A lot of your posts make me feel like I’m missing a lot of very specific context.

No worries! Didn’t take me more than a couple of minutes to find. I just posted the link in case anyone else would want it.

Found it: https://live.acarsdrama.com/@acarsdrama/116708457056507627