Are we putting ourselves at a competitive disadvantage to big tech et. al. by being anti ai purists?

It doesn’t have to be either/or.

You can use LLMs as a tool and never touch Big Tech. You can use them for utilitarian purposes, without getting all cultish about them. You can actively hate “AI” while doing this.

Honestly, I think Big Tech is stoking the “Pro AI” and “Anti AI” divide, because what they’re most scared of is the public viewing it as yet another dumb automation tool. That would obliterate their business. They want the public to see it as an “all-in-with-us-or-nothing” affair, kinda like they did with social media.

A couple of things:

• Trashy western sex appeal (or tacticool fetish, drug glorification, or whatever) is a fair point. That’s a problem. Some of those trashy western shows are popular, too.

• But I don’t buy whataboutism. It doesn’t make either thing okay.

• Some popular anime really takes things to an… extreme.

• And even outside the notorious anime, I kept finding problematic tropes when I tried to get into anime, like these women written in a way that got me to squirm in my seat and turn off my TV. When I sample critically acclaimed western TV/animation, this is not such an issue; I don’t mean to make a comparison, but just to convey that anime sampling made me uncharacteristically uncomfortable.

I hear this and sympathize.

…But. Honestly? It’s not just western fans that are the problem. A few anime shows I researched seem to use underage sex appeal (or other problematic depictions, like toxic relationships) to draw attention, to make money.

At some point that’s not a cultural difference or misunderstanding; it’s pandering. The showrunners know what they’re doing. I’d argue that is exploitation of Japanese culture.

And even if they’re rejected by fans like you, these problematic shows seem to be immensely popular.

Money?

IDK, I guess we will just disagree, as I feel like I’m reading a completely separate blog than everyone else.

As the other read makes zero sense. A model cannot be “protected” because it cannot be damaged, it is frozen in place, its state does not change. Anthropic seems to be talking about people “abusing its models” for harmful output, but people here have read it the other way just because of English language ambiguity.

+1

Honestly this should ship with many browsers, be a default for search engines and such.

Your confusion is justified. It shouldn’t work, all things equal.

I think it’s because the intake of a plane is more diffuse; it’s pulling air in from different directions. Meanwhile, the “reverse exhaust” is ejected from a smaller nozzle at higher temperature; it has a higher velocity, therefore more momentum to “push” the plane than the intake “pulling” it.

For the record, there’s a similar scheme for sailboats. Hook a fan in the air to a propeller in the water, and the wind imparts less momentum than the prop “pushing” on the water. Such a “rotary sailboat” can sail straight into the wind.

Another factor is that model capabilities are topping out, but inference costs of cheaper models are absolutely plummeting.

In other words, it’s a race to the bottom. And inference will become a dirt cheap commodity.

…And what do we need all those datacenters for if inference is dirt cheap?

They’ve been around (and somewhat notorious) for a long time, even before “AI” became a popular term.

Theoretically.

In practice, 95% of consumer GPUs are too busy and too small to run an LLM good enough for dialogue, or a decision model smart enough to be warranted over hardcoded logic.

CPUs are too slow. NPUs aren’t universal and are, unfortunately, all but useless with modern software frameworks.

I am very much for LLM NPCs; there are some fantastic mods for such things.

But in shipping games? That can’t be jank?

It is not time. And I would never buy a game with a hard cloud dependency for SP. We will have to get past the RAMpocalypse, at the very least.

I personally stay away from AI due to the many issues AI brings.

I am a fan of people fighting big companies.

I have had an influx of these game decomps on my Instagram recently.

I am not trying to be pedantic, but Instagram is “AI.” It was toxic AI before LLMs were a popular thing. Participating in their algorithm is basically everything dystopian about the world of machine learning, concentrated in one service. And using it helps Meta perpetuate it more than using any LLM does.

Again, I don’t mean to be hostile; maybe you use it sparingly in some cleaner way, and I’m not saying everything on Instagram is bad; quite the contrary.

But I just want to make what Instagram runs it crystal clear.

Well… Our points of view are so far apart, I don’t even know where to begin to reconcile them.

Sourcing, information hygiene, proof, and science are irrefutably central to the way I wish to interact with the world. It should be the bedrock of any healthy online community structure, IMO.

Places where that’s not enforced are place people just echo what they want to hear, in my experience. Whether its true or not is secondary, as apparently information hygiene is too “heirarchal” and constricting.

I don’t want that. I don’t want anything to do with the Fediverse if that’s what it becomes, and it sure feels like many users want that.

…you’d be impressed at my fortitude…

Again, respectfully, this kind of argument is not going to win you sympathy.

I will point to the golden rule of the internet. If you are frustrated with other commenters:

Don’t feed the trolls.

You don’t have to engage in such threads.

TBH you are not making a strong case for yourself here, OP.

It sounds like the thread was locked because the discussion got toxic, which indeed takes multiple participants.

But your presentation of your question, solely based on your own description and setting the actual topic aside seems… abrasive. And the replies here seem abrasive.

I mean, you shouldn’t. “XiJinpingStanAccount” is a glorious handle. I salute you, comrade.

One I’ve had, that’s adjacent, is allowing communities to form “hierarchies.”

So niches could be under /c/technology, individual video games or genres could be under /c/gaming, sports, you name it. It would all require consent from the lower and upper parts of the chain.

One benefit would be “pooling” housekeeping moderation. e.g. if a troll harassed you on a niche community, one from a higher one could moderate it quicker. Or if /c/technology was overwhelmed at a point in time, smaller communities mods could step in to housekeep comments in /c/technology, with the granularity of all this being configurable.

I bring this up because I think it helps the issue of moderators not being responsive enough to your comments. And those passionate enough to run some niche community well enough to fit in a hierarchy would earn the power to moderate comments under topics they’d frequently post to anyway. But it has a lot of other benefits, like helping with discoverability of smaller communities without an algo.

Let’s perform a thought experiment.

I disagree with YOUR post. This one, right here, and I really try to politely articulate my point as to why.

You don’t like it, so you delete it. You delete any post with similar sentiment.

…Is that fair?

If you response is “well, I wouldnt do that,” think about it. That’s the power you’re asking every user on Lemmy user to be responsible with. Do you trust every user to that extent?

deleted by creator

Eh. I still (respectfully) disagree.

I think that’s newsworthy; but it should have linked the dev announcement, or GitHub thread, or wherever the information was sourced from.

If /c/technology won’t allow direct posts from developers, that is indeed a problem for /c/technology, as that’s a first party source for topical news.

But to make my position clear:

• If that poster was intended to post without a source, that should be against the rules. It’s how I would moderate /c/technology.

• There are other communities to post that to.

• Hence, I don’t agree that this issue should be generalized. Communities need to straighten out sourcing rules to be more practical, especially to allow topical first party sources, but I don’t think they should loosen sourcing rules.

If Lemmy/Piefed has discoverability issues for small communities or an under moderation issues, I think those are structural, and we should change how the site works to address them.

That’s what I was thinking, TBH. I have been scared for his life, but I am more scared now.