PLEASE be extra careful of YouTube videos you post here. There's a tsunami of misinformation, much of it AI slop.

[highplainsdem]

You'll see lots of fake news about encounters and experiences Trump and other politicians and celebrities had. These are often obviously AI, but some people still can't recognize AI art and video.

You'll see videos from YouTube channels making political points and claiming to be sympathetic anonymous people. Yesterday one example was a channel purportedly from an elderly man who was dying and was trying to raise the alarm about the decline of British society. "He" was also criticizing both the ruling Labour Party and the Tories and praising Nigel Farage's Reform Party. That supposedly dying man was generating multiple videos with AI slop per day, trying to get people upset with the UK government the way Fox News riles up their viewers and offers Republicans as saviors.

Be wary of any YouTubers whose activity elsewhere online you can't verify. Reliable news outlets, pundits, podcasters and journalists are fine. Watch out for newer YouTube channels posting a lot of videos, especially multiple videos a day.

If they use AI to create art, they're more likely to use AI to write scripts and have them narrated by AI voices.

If they use AI to create satirical or parody videos, even if those videos attack Trump and other people we despise, they aren't necessarily liberal. They can very well be equal-opportunity clickbait producers, since this stuff requires little or no effort. Check their older videos. Sometimes with channels of that type, you'll see shifts in topics they focus on, testing different types of AI-generated clickbait till they find the best one to reel in viewers. They don't deserve views from DUers.

I wish it weren't necessary to warn people to watch out for this sort of garbage online, but generative AI tools have made the internet much worse in just a few years.

[bucolic_frolic]

1. I've encountered a few non-political ones too, on other subjects

and even though they stray from their subject, and never make the point they said they would, the public still praises them. The stupidity cannot be overestimated.

[Ocelot II]

5. There's good stuff on YouTube, but like every other social media platform, you have to pay attention

to where the material is coming from. Anybody can post anything; some of it is AI slop and some of it is just slop. Pay attention to the source.

[highplainsdem]

6. I agree there's a lot of great stuff on YouTube. But the tsunami of AI slop is making it harder to

find the good stuff, and probably taking views and revenue away from what's worthwhile.

[spinbaby]

51. And it's such BAD AI slop

So much is obviously AI if you’re paying the least bit of attention, but somehow people aren’t noticing. I recently came across a video about the reintroduction of European buffalo in Romania, which appeared to have been assembled entirely by AI. It not only featured many clips of American buffalo, but also a bear and a chimpanzee, all of which apparently were close enough for the AI.

[Ocelot II]

3. In particular, watch out for confirmation bias.

Everyone is susceptible to it. We want to believe bad things about our adversaries (often with good reason), but not all of those juicy bits are actually true. Sometimes it's satire; sometimes it's just people stirring the pot - and AI slop makes this so much easier. There was just a post about some nasty thing Trump was supposed to have said about the Buckingham Palace guards, but which couldn't have happened because he was in Scotland and nowhere near Buckingham Palace. It was sort of believable because Trump is a crude asshole who really might have said such a thing, but it was total AI crap that was posted on Youtube without credible attribution. ALWAYS CHECK YOUR SOURCE BEFORE POSTING, especially if it's something outrageous that you would really like to believe. AI is making an already fraught political atmosphere that much worse, but we don't have to be part of it.

[highplainsdem]

8. I posted the AI slop video the OP in that thread was apparently referring to, in a reply there, and the YouTube

search results showed three similar but slightly different AI slop videos from different accounts set up for AI slop at about the same time.

[mahatmakanejeeves]

29. I saw that post. NT

[WarGamer]

35. aka catnip.

[Hekate]

42. This. Absolutely this.

[Celerity]

4. The same for X, Reddit, TikTok, etc, and even some Bluesky posts, plus many rando blogs and dodgy sites.

[Ocelot II]

7. Yup. There's no filter, and people post crap everywhere - AI or just regular crap.

It's more important than ever to be discerning and not assume it's true just because you saw it on the Internet.

[Brenda]

24. Agree. And maybe it's not just innocent or ignorant folks posting these threads. nt

[FakeNoose]

9. We should have a review committee to vet these vidoes before they can be posted

I agree that there are way too many videos. Also the quality of some (few) of them is questionable.
Most of all it's whether they are truthful or not.

If a review committee - a group of knowledgeable DUers of course - thinks the video is worthy, then it can be posted. But a lot of them wouldn't make the grade, in my opinion.

[Wednesdays]

36. That would be good only IF there's a way to go through that process within a few minutes

Let's say there's a REALLY important breaking story. We'd want to see the video about it right away.

[FakeNoose]

37. OK ... maybe they only need to be vetted if they're from something other than a trusted national news source

I'd agree that a video that was originally published by one of the major news channels wouldn't be a problem.
Those stories aren't what's causing the trouble anyway. We KNOW YouTube isn't doing anything.

[littlemissmartypants]

43. We do have alerts but we need something more, imo. ❤️

[hunter]

50. I just don't watch political videos.

There's little reward in them for me. They seem mostly noise.

Other than the Lounge, I skip posts with the little red square that marks YouTube videos. I skip all posts with the little blue square that mark the site formerly known as twitter.

[FakeNoose]

52. Right, I agree ....

However some people (who are less discerning than we are) might see the misinformation and believe it. They would think, well it's on DU so it must be true. We shouldn't allow that to happen.

People who can't tell the fake news from real news, well they're the ones who vote R and watch Faux Noise and ThikThok.

[SheltieLover]

10. Kick

[blogslug]

11. Here's some things to look for

1. The name of the Youtube channel. Never heard of it? Probably bogus

2. Look at the Thumbnail/preview image provided. Is everyone very smooth and the colors extra saturated? Probably AI

[BidenRocks]

18. I've searched the googles

I found my vid was produced by a Chinese company, and they showed the other names they use and links to YouTube.

Usually there are repeat themes and mispronounced names, places or things. They also love adjectives as much as chump.

I watched for weeks and finally caught on to the pattern.

[NewHendoLib]

12. Same with Instagram reels. Often the voice is a giveaway.

Even if not AI, just so much disinfo (garbage gardening tips)

[Gimpyknee]

13. You mean those Himalaya salt ads promising weight loss are AI generated?

That’s not Ophra Winfrey?

[mgardener]

23. Damn

I live in hope of a fast weight loss plan!

[anciano]

14. Good suggestion .....

IMO, online discussion forums should be for original and genuine "discussion", not just somewhere to post someone else's tweet or video. Just saying....

[J_William_Ryan]

15. "PLEASE be extra careful of YouTube videos you post here..."

It would be nice if no YouTube videos were posted here.

Unfortunately, that’s not going to happen.

As correctly noted: most are suspect, consist of confirmation bias, and offer very little in useful information.

[hunter]

53. Sometimes a video is the most effective way to describe something.

I posted a video just today about modern olive harvesting equipment as it relates to farm labor issues. I regularly watch animal videos in the Lounge.

It's the news, politics, and opinion videos I have no tolerance for, AI "slop" or not. I don't watch any of that stuff on television and I don't watch it on the internet.

[Fiendish Thingy]

16. A lie travels around the world before the truth can get its pants on

Critical thinking skills will an essential tool for the survival of democracy.

Thanks for the post.

[IcyPeas]

17. The pronunciation is often another clue. AI mispronounces words. Example

In that video purportedly of Trump and the guards....

The voice over says Prince William told the guards to "resumé your post"

Pronounced it Resumé as in C.V. as in bio. Instead of resume as in continue.

(It was at the 2:18 mark of that video but that thread was locked by the time i wanted to post)

I hate these AI voice over videos too. That "channel" is full of fakes.

Thanks highplainsdem for taking the time to fact-check these videos.

And at least the OP of that post deleted it. Some posters don't even take down a video or photo when they are told numerous times in a thread that it's fake. And then there are posters who don't read the previous posts saying something is fake and keep replying to the OP.

Oy! people! We can do better.

[calimary]

19. "Better safe than sorry" - boy THERE'S one that remains relevant!

[LiberalArkie]

20. Are you saying the "live action video" of Trump running naked through the desert is not real?

[Kaleva]

21. New vids narrated by Christopher Lee

[NGeorgian]

22. The youtube I just posted I first saw on MSNBC.

[Amaryllis]

25. I have to say I could use a tutorial in how to recognize AI. Sometimes i can tell but I don't know enough to recognize

it all the time.

[WiVoter]

26. I Prefer Articles Wirh Text

I don’t have a lot of time to watch videos, when I can read an article much quicker.

[Marcuse]

27. An authoritative YouTube AI gleefully describes Cabot's cuckoldry and details his financial preparations for divorce.

Another notes that he has been divorced for three years. All accompanied by slick B roll.

[highplainsdem]

48. There are "authoritative" AI videos with obituaries of people who are still alive, sensational accounts of

their last words, etc. And while there were some fakes like this before generative AI tools became available, there are a lot more now.

[FraDon]

28. How do we not already have AI filters,

that instantly identify and display what one is seeing? Seems like a very ai kind of thing.An ai Turing test?

[highplainsdem]

49. AI detectors are notoriously unreliable. There are some platforms using different techniques to label

AI images and video generated by their own AI tools, but not only do those work only for their own AI slop, but there's no certainty those ways of tagging something can't be removed - because people will be trying to find ways to get rid of them.

Just as the people behind popular AI tool Midjourney - notorious for spitting out AI images containing one or more watermark from the billions of images the company stole for training data - worked to keep that pesky evidence of theft from showing up.

I vaguely recall something about OpenAI admitting a year or so ago that they would be reluctant to make their AI tools' output always identifiable as AI because their USERS don't want it identifiable as AI.

[mahatmakanejeeves]

30. I can think of two recent threads in LBN that are deceptive.

In one thread, the incident that was said to have occurred never happened. This was clearly established by video of a White House presser at which nothing dramatic happens, despite the claim. The “news” source was a tabloid. The story’s author made the whole thing up. The poster was informed of the false claim, but he didn’t care. The OP got something like 155 recommendations.

In the other thread, the source was a British paper that lives for clickbait. It was trying to gin up clicks for an incident it claimed had just occurred a day or two earlier. The incident had occurred, but a month earlier. The source had not checked the date of the incident. It never retracted the story. As far as I know, the thread is still in place.

[DownriverDem]

32. Here are some on youtube (free one) I trust & like

MeidasTouch, OccupyDemocrats, The Romano Report, Keith Edwards, Luke Beasley, Brian Tyler Cohen, Adam Lockler

[msongs]

33. the value of the content is usually inversely propotional to the amount of caps nt

[Srkdqltr]

54. Yes. ALL CAPS IS A GIVEAWAY.

[defacto7]

34. Wise advice. n/t

I imagine AI won't always be as stupid as it is now and that is troubling. Our work is going to get harder.

[Waywardpedestrian]

38. You Tube/Videos

I prefer to read an article, essay, substack etc. The onslaught of liberal you tubers and assorted videos has changed me regarding my necessity of current events. I grew up reading the Chicago Tribune (historically a nasty lot) and the Chicago Sun-Times. I knew what policies each of these two daily newspapers were trying to sway their readers to adhere to. At the very least both sides were represented even when it was obvious the article in question would be a pathetic attempt at defending the Democratic opinion on these issues. Those days are long gone. Where can anyone find articles of depth, interest and relevance to all the horrors perpetrated by the current administration?

[Jack Valentino]

39. 99.9 percent of the Youtube videos I ever post here are rock music performances...

do I get a "pass" ?

[Waywardpedestrian]

40. You Tube

Absolutely! Let us hear.

[Jack Valentino]

41. "The Waiting is the hardest part"--- Never more than now

Tom Petty, gone much too soon


His passing hurt me more than any other artist in the past 10 or 15 years.....

[littlemissmartypants]

45. Sources need verification. Clicks make money. So...

For some it's a game. You Tube as a source just because it's music shouldn't get a pass, imo.

[highplainsdem]

47. See reply 46.

[highplainsdem]

46. I host Music Appreciation, and almost every OP there has at least one video with music. Those are fine, unless

they're made with generative AI, which I've asked people not to post there because it isn't real creativity, and it hurts real artists. Unfortunately there's already enough AI slop flooding YouTube and music platforms like Spotify that these days I check anything there I'm not already familiar with. And TBH, though I love seeing new music and artists discovered, it's sometimes a relief to see that something was uploaded to YouTube more than a couple of years ago, before the AI slop started. The biggest problem.I've seen with AI slop there is people being tricked by new AI-generated channels producing anti-Trump songs, with those YouTubers often having tried different subjects as clickbait first.

What I'm.talking about here, in this General Discussion thread, are mostly videos purporting to be real news or commentary. The "real news" at least can be checked by seeing if any real news outlets have the story. Commentary being offered allegedly by people who might not even exist - might be created to seem like a sympathetic narrator - can be trickier to catch. But if there's AI art (especially AI pictures of the alleged creator that vary from one video to the next) that's a good clue, as are multiple videos in too short a time. And honestly, there are plenty of credible real people using YouTube for commentary - more than most DUers could keep up with now - so there's no need to find new "people" on YouTube offering commentary when their real intent might be making money off the platform. Which AI makes all too easy.

[littlemissmartypants]

44. Can we use an alert? What rules do we have regarding mis/disinformation?

If someone knows it's misinformation and shares it (-D-) deliberately it becomes disinformation. To me that deliberate sharing should earn a stiffer penalty that just posted stuff without vetting.

I don't think we really have any clear guidelines on the problem. We probably new a focus group, some team education, some firm guidelines, possibly a special rule. Otherwise, the lack of clarity won't help and feelings are going to be hurt. Something that may sound inconsequential but we can't continue to build a strong community here without empathy and respect.

(Plus, I'm an INFJ so it's my weakness/superpower.)

Anyway it's a big problem which is only going to expand if we don't have guidelines.

I love DU. I don't want it to suck. It could possibly if the junk content takes over.

Everything has a tipping point.

We can all try to do our best individually but there really is strength in numbers. Especially when those numbers are as intelligent and diverse as DU.

I think we need some guidelines.

Thanks for bringing this up.


❤️ pants
RESIST!! ✊️

[LiberalLovinLug]

55. I think it would be more useful to dedicate an area or forum that DUers can post suspicious videos from YT

To look out for those ones. And know they are fake.

And just to show others what they look like.

Ignoring the problem won't make it go away

[littlemissmartypants]

56. I thought the idea was to curb the unsubstantiated content...

Not signal boost it in a special place?

Maybe I'm confused.

[LiberalLovinLug]

59. I think it would be interesting to have a forum to post and debate how fake, or not, a video is

and to post them as a caution of which ones to watch out for.

I highly doubt any DUer, especially if told in advance that the videos posted are suspicous to believe, would actually be mislead by anything posted in that context. If that's what you are worried about.

Its not about re-tweeting them, giving them more life, its about having an in-house corral to study them, and stop anyone signal boosting those false ones, who may have thought they are real.

And fake ones that are 'too good to be true' from our point of view can be suspect as well. And dangerous, politically. Nip those in the bud. As Democrats and supporters are 1000x more in trouble for spreading something later to be debunked ie. Dan Rather.

[highplainsdem]

60. I realize you mean well, but there are hundreds, maybe thousands, of such videos added to YouTube

every day. It would be a terrible waste of DU to post them here just as examples of this garbage, and it would also give them more views. But it's important to identify the ones DUers have been tricked by when some do get posted here.

[Montauk6]

57. I've also grown WEARY of the clickbaity thumbnails and titles.

"TRUMP LOSES IT"
"TRUMP HAS A MENTAL BREAKDOWN"

And it's usually something he posted on his Fibs Sociopath webshite instead of what the thumbnail illustrates. So, I've been ignoring a lot of them. Don't get me wrong, many are right on but, again, I can't stand lures.

[LiberalLovinLug]

61. I hear you

MediasTouch is a prime example. Even though their content is watchable, and important, and they have good hosts. So annoying. They don't need to do that. I don't know who their marketing person is but please stop.

[Martin68]

58. I've criticized several posting, but risk being blocked whenever I do.