Maybe not that interesting for everyone here, but I found no better community for this.
I want an ad / DNS blocklist for AI slop. I want to not even be able to go to HiSToRiAn AsLeEp In ThE WoOdS channels on my networks and devices
You can have your Google search go before 2022. I do this sometimes
Google still does its psychotic summarys though.
Then other day I was trying to find out how to get just one AD account not to sync through to Azure without disabling it. And the AI came up with this complicated instruction set that didn’t work and was totally made up from nowhere. Now what on earth was the point in doing that?
At one point it told me I had to triple click on something. Because that’s totally a thing.
It’s so fun when it’s so specific about some detail with casual confidence that is based on absolutely nothing at all. I know ultimately it’s architecture is more akin to a predictive word generator, but it seems so much better.
Saw a clear demonstration and it is wild that the output is consistent, but at least in the model I saw being run, every word is generated without it having considered what the word after would be or what the general concept it is going for. For a human one has to already know the concept before he/she starts putting words to it, but at least the models I’ve seen explained with detail, it manages to assemble it word by word without knowing where it is trying to go in advance.
Yeah, after seeing it hallucinate on things I actually know, I’ve decided to never trust it.
You might try setting Google’s Web subsearch as your main search engine, or else put your searches through https://udm14.com/, which does the same thing.
Honestly, it isn’t the AI so much as it is the suspension that rules that apply to us should also apply to AI. They are barely legal companies getting away with murder due to the residual power of a corrupt empire in decline.
Much the same as blockchain - it could have been exciting and fun, but it was immediately put to work to exploit people, and in the end that’s what the tools effectively were.
I don’t think blockchain has any legitimate uses. All of the proposed uses for blockchain are all tech bro nonsense.
I think they were talking about having a decentralised property permits. As if that’s something that would be even remotely useful.
You’d be surprised how valuable these technologies can be when they’re not only being developed to create new financialized services to overcharge people for. Manufacturing and shipping has huge capital and logistic buffers created by information gaps, production shortfalls, fraud, etcetera, and cryptographic ledgers are excellent for ensuring accurate production and inventory information is relayed to clients in realtime. What’s more, it has the potential to mediate international transactions without relying on Swift or global banking institutions that are an extension of US authority.
Notice how nothing I said was about fucking NFTs. The western financial system completely lost the plot doing super cocaine and became convinced they could just ask the rest of the world to stay slaves forever. They don’t even care about its potential! They want TrumpCoin.
Yeah, and if it weren’t for the techbro nonsense, it would be tossed onto the pile of mathematical curios that don’t have nontrivial uses.
The thing is - we often do find uses for those curiosities years later.
In the mean time, I wouldn’t mind if a decentralized video game came along where game assets were decentralized, distributed by bittorrent, and player assets were decentralized and tracked by blockchain.
I have noticed that I don’t click on their videos as fast. Used to be that as soon as you see a video in the feed it was an instant click.
I think the topics they make videos on these days feel more gimicy and click baity so I kinda avoid them
100%. The recent videos all seem like they’re ragebait (‘alcohol is awesome’ springs to mind).
They could be good, I haven’t watched them. if they’re targeting people who respond to that tone then it’s not for me.
It’s a shame. I liked the earlier videos on ant colonies and strange matter and stars and stuff.
Yeah, me too. This was instant click though.
And instant ad for their merch.
I get the frustration, but it does actually cost money to make content, especially high effort content like this.
Add on top of that the fact that the people involved need to be able to eat, have a home, provide for their families, have a life.
People would complain if they were sponsored by some shitty VPN provider or the like, and also complain about them trying to sell merch. I certainly wouldn’t work for free, so I don’t see why they should have to.
It’s not really that hard to do a couple of key presses to skip ahead, either. It’s what I do.
Honestly, I don’t care. They chose to “make content”. If they aren’t making enough, get a real job.
Kurzgesagt is annoyingly techno-solutionist but now that they are a big target of AI slop (their artistic style being heavily copied), they complain. That’s good I guess.
If you actually watch the video that’s not what they’re talking about. They are talking about how it’s difficult to fact check information in a world of AI misinformation.
I always find it’s easier to know what a video is about if I actually watched the video. It’s just this little life hack I’ve come up with.
Watch the video? Duh. Thanks I watched it, and I know they don’t talk about how AI steals their art style (even though it’s really happening). I just pointed out the irony. The video also says they are usually enthusiastic about new tech but not this time.
I know this channel well enough to have seen countless videos praising fake tech like carbon capture and that’s why I have a problem with them. Am I allowed to have an opinion?
You are allowed to have an opinion but when it’s a stupid opinion expect to be ridiculed for it.
Carbon capture is an actual technology it’s real and it works. There’s not a lot of it about yet but there’s also not a lot of fusion reactors yet but they’re also real technology.
Techno-solutionist lol. The power of labels eh?
What is the problem with techno-solutionism? Is it just that under capitalism techno-solutionism often results in corruption? Is the development of lemmy not techno-solutionism (to the enshittification of reddit)?
Have they considered that they’re just not a great channel?
Have you actually watched any of their videos? That’s obviously not true
I dont think its AI Slop thats “killing the channel” if that statement is even remotely true its due to more and more of their videos being slop sci low information videos. The target audience must be stoners who want to watch something that is visually appealing and feels educational
For what it’s worth, I kinda agree. Maybe I’ve changed but I feel like their content over the last year or two is nowhere near as good as it was.
Maybe I’ve just had a change in taste, though.
They’re fine. Jesus christ.
Kurzgezagt is a slop channel too you know.
You can’t really call it slop just because you disagree with their views and representations of things.
Their stuff is carefully researched and sourced, human crafted and open to critique. Whether they’re correct in their assessments or not is of course up for debate, but it’s good craftsmanship and they show their work.
Careful research ?
Careful research.
Eh, there’s a lot of blending of conjecture, opinion and fact all presented as truth, and their handling of mistakes could be better - they’ve openly said if they consider a mistake to be minor then they don’t even issue a correction or update.
I personally think that attitude towards production pushes it towards slop, as for things like entertainment one of the key defining things that separate slop from quality media is passion, but if you don’t care about making accurate content then are you any better than just getting AI to write a script?
Tell me you live in an internet echo chamber without directly saying it…
If echo chamber means I can criticize your gurus, then so be it
And of course, the video was preempted by an AI slop ad of a talking monkey wearing a neck brace.
!videos@lemmy.world is a good one for stuff like this. That community loves their work.
world no thank you
Boo!
I watched this with my son last night. Quite enjoyed it. I find the cadences of the narrator’s voice oddly soothing, although ironically it sounds very like an AI voice.
I’ve been watching that channel for years it’s very high quality.
I mean there’s been some controversy surrounding a number of their videos. Some were under fire for poor research or demonstrating a singular, not widely-held view on certain topics. And I think for one video they were accused of plagiarism iirc. This was when it was still somewhat early days for the channel, haven’t followed them since so not sure what the state is now.
everything has controversy. I really do not care
In the video you’re literally commenting on they explain how they fact check their videos and how they work with professionals in the field at matter.
You can always complain that something wasn’t as good as you personally wanted it to be but that may very well be bias. The people making the accusations may have a bias. All we can do is take the totality of their output and cast judgement on that.
The point was that they haven’t always held themselves up to those standards and have sometimes only used professionals espousing a single viewpoint (where multiple exist).
I should mention this isn’t bias, iirc the channel did release a video apologizing for some of the issues (though not all), so it wasn’t even up to their own standards by their own admission.
There’s a wikipedia entry listing some of the controversies: https://en.wikipedia.org/wiki/Kurzgesagt#%3A~%3Atext=While+some+commentators+have+praised%2Cand+use+of+emotive+language.
Looking things up now, I see that the plagiarism case was slightly different: they had published a video on addiction, which was fairly explosive in its claims. Turns out it was citing basically just one fringe researcher who was also accused of plagiarism. The claims did not seem to hold up to scrutiny.
When another channel doing a series on how pop-sci influencers can sometimes spread misinformed ideas asked some questions to Kurzgesagt, they were immediately a bit apprehensive but agreed to do some interview questions, though with the caveat that they were busy with other things and needed a few weeks before it could take place. Then before the interview took place they suddenly put out their own apology video and took the addiction video down. At no point was it mentioned that another channel prompted this action, it was presented as some kind of inward reflection that they had come to themselves.
What do you think AI voices were trained off of?
I feel like the title doesn’t match the content.
The video gives an elaborate description on their evaluation of “AI” and it’s influence on the Internet at large. And then they conclude with “we’ll continue like before” directly contradicting the title.
Feels disingenuous. And ironic after they talked about their extensive investments into fact checking.
The video gives an elaborate description on their evaluation of “AI” and it’s influence on the Internet at large. And then they conclude with “we’ll continue like before” directly contradicting the title.
You missed the entire point of the video.
The claims are simple:
-
in order to make this type of videos, they need to be able to reliable fact check
-
data on the internet is increasingly polluted by AI slop, making it harder to distinguish fact from slop
-
for now, they have no choice but to continue while being extra vigilant… but eventually, if things do not change, they will be unable to perform
It’s the exact same situation about climate change… we need to act now, most of us will suffer otherwise but for now we continue on living while trying to adjust where we can (recycling, reusing, less/no meat, etc) even if we know that will not be enough long term.
Also, presuming they are sincere and put in all that effort, they are competing with other sources that have no such discipline and they are able to flood the field and grab eyeballs faster than they could.
exactly, also explainedein the vid
Not the guy you’re answering to, but I kind of agree with him, the point is fuzzy and the title is clickbaity. With sucha title I expected they would present numbers and figures.
well, it may be a matter of context and tolerance here but I find the concept they are presenting is axiomatic and as such would not require any further explanation:
They use the internet to research their videos… the internet is getting more and more polluted with false narratives… ergo, it is becoming harder to research for their videos. Without good source, there are no videos.
If I tell you plants need water to exist but each season brings less and less rain year after year… would you say a title such as “drought is killing the plants” clickbaity?
I assumed they don’t do their research using random crap on “the internet”, but reliable experts, peer reviewed papers and such. No specific claims about topics, funding, time or anything. And again, no numbers, so hard to argue objectively.
I assumed they don’t do their research using random crap on “the internet”, but reliable experts, peer reviewed papers and such
Yes, that is what they claim. But I am sure you have seen how hard it is now to find something even if you know exactly what you are looking for. It’s not like there are 2 libraries online for anything you need, right? You start researching about topic A and read that Dr XYZ did a study on this so you look for that study… just to find out Dr XYZ does not and has never existed.
No specific claims about topics, funding, time or anything. And again, no numbers, so hard to argue objectively.
So you want a specific number as to how many bad sources they are now forcing to discard because they turned out to be AI slop?
To begin with. Time delta. Cost creep. Anything.
Those metrics aren’t any more trustworthy than their own subjective word anyway. If they wanted to say they took more time then they could delay at their whim anyway. If they said their production costs increased, then again, they could spend the money to fit the narrative. On those particular points objective evidence is so susceptible to being gamed that it isn’t really more valuable than their subjective reporting.
Numbers of subscribers/views could be a bit more informative, but then people inclined to disbelieve would claim it’s because of any number of other reasons not because of AI slop.
it’s not that type of channel… they never do more than a percentage or a rate.
their thing is to explain concepts in a way a young audience can digest them
Killing in this case sounds like the content is becoming harder and harder to create, which they lay out the subjective case for, but that wouldn’t be exactly something they could use figures to present, since it’s so subjective.
The one point they might have been able to show with numbers would be the emergence of AI slop ‘infotainment animations’ diluting the audience, but that wasn’t exactly the biggest point of the video and it might be a bit early to be able to demonstrate statistically credible evidence on that one.
-
DeArrow shows the title as something like “why you can’t trust AI with facts”
Do you think if we pool every AI in the world it will be able to figure out the difference between its and it’s? Seems unlikely.
The channel hat always been disingenuous. It’s not the first video they have where they develop a well written essay that has conclusions that make no sense with the information presented. It’s the theater of research without any of the substance. The editors just do whatever they want, under the expectations that the writing team will support their preconceived notion.
They’re an entertainment channel, not a science communication channel. They have said some awful, totally not fact supported stuff in the past.
They have said some awful, totally not fact supported stuff in the past.
Like?
Climate change whitewashing for corporations with awful conflicts of interest. Others have posted the links to the videos elsewhere in this comment section.
Wow look at all of the evidence you’ve provided. It’s going to take me all night to go through it.
If you going to make claims like that you’re going to need to provide even a shred of evidence
As far as science channels go, you’ve got SpaceTime for college students, Veritasium for high schoolers, and Kurzgesagt for newborn infants or maybe a smart dog. It’s probably at about the right level if you want to explain science to an Australian Shepherd.
c/iamverysmart
Veritasium was bought years ago. No editorial freedom. Never heard of SpaceTime. I have better sources for science than YouTube slop, thank you.
Given the fact that you don’t appear to know how to cite your sources I’m not convinced that you in fact do have better sources for science.
Awww is it the first youtube video you watched?
Did you come into a comment section and expected not to see any comments?
Do you take everything as it is, without criticizing anything?
Do that if you want. No need to be so dismissive without actually making your point. Which I assume is that clickbait is “normal”.
It’s definitely not only that.
I and, from what I’ve seen discussed here and elsewhere online, many other people have stopped watching the channel because of many valid reasons completely unrelated to AI. I haven’t watched a single one of their videos for years, something changed in their content and I just lost interest. Then there have been several controversial events throughout the years, both objective bad stuff and subjective things that made many loose interest and faith in their integrity. They definitely became one of the many channels that lost it’s way because of how big it got. The animations became too “perfect” in a weird way, they lost their personality and they also got scared of having real opinions so they started doing this “all sides” shit and that’s when I tapped out. I’ve tried to watch new videos from them about once every 6 months or so, but I can’t even make it past a minute without completely loosing interest.Completely agree, their channel has changed a lot and seems to be producing videos on a conveyor belt now, while before they used to do one video or maximum two a month. Now it seems they produce a video a week, and interesting topics are more hard to come by.
When they said that they’re “almost 70 full time people and a lot of freelancers on top” I almost did a spit take. I know there are big channels and operations on YT, but this seems such an unreasonable amount of employees for this type of channel and audience. No wonder it feels oversaturated and overdone, they probably feel the need to put more and more videos to keep their huge team and expences afloat.
Just find a sustainable pace and team size, don’t go the corporate way of growth over all.
You didn’t even watch this video.
In that it’s mostly a merch ad hidden behind a clickbait title.
So I guess it’s a good test for that sort of “just read the headline” response.
It’s been a rough few days and I think I may be coming around. What hope is there to parse AI misinformation if people can’t parse a Reddit-like link aggregator?
I may be done with this place at this point. It’s just all bad. If not the whole Internet, certainly the whole of social media.
This is not an airport, you don’t have to announce your departure
I think you might be referenced in this video 😉
This whole video felt like an advertisement to me. I’d already seen it twice, was it necessary to crosspost it again? You could at least have waited a couple of weeks before reposting the same content.
Kurzgesagt became (non AI) slop itself so im not surprised. All the videos of them that i have watched in the past few years were disappointments.
Ironic coming from one of the preeminent traditional slop channels
They use traditional animation for all their videos, what are you talking about?
Yes, that’s why I said traditional slop, not ai slop.
What’s sloppy about Kurz? I don’t get it.
Oh, my bad. Completely glossed over the word and was too focused on slop relating to AI.
This video is very probably also AI slop
Have you ever watched the channel? They even made a video outlining their process which takes months of work.
Nope, and this one is NOT going to change that … can’t take all that cutesy animation and the fast-flowing babble for 2 minutes, let alone 12
Imagine being this proud of ignorance. I can’t wrap my head around it.
Or they just don’t like the way Kurzgesagt presents their information. Just like they wrote. Different folks, different strokes.
I didn’t say they have to like it, that would be silly.
I’m criticising them for making an incorrect statement, being corrected on it, then acting extremely proud of being ignorant of the facts, and committing to not informing themselves.
