Latest / Mini Scroll / John Pork Isn’t Dead, But The Internet Might Be: Why AI Slop Has Taken Over Our Feeds

Transcript

John Pork Isn’t Dead, But The Internet Might Be: Why AI Slop Has Taken Over Our Feeds · Mini Scroll

• 0:00
  This episode of Infinite Scroll is brought to you by Britta from
• 0:04
  Water Talk to the Emotional Support Water Bottle.
• 0:06
  Gen. Z and Millennials have developed
• 0:09
  an obsession with water and staying hydrated, so if you're
• 0:12
  ready to take your hydration game to the next level, Britta
• 0:15
  is the way to go. With Britta's Water Filter Jugs,
• 0:18
  you can enjoy pure and fresh tasting filtered water straight
• 0:22
  from any tap. By ditching single use plastic
• 0:25
  bottles, you'll be doing your bit for the planet and drinking
• 0:28
  water that tastes amazing. Britta doesn't just filter
• 0:31
  water, it filters out the guilt of plastic waste.
• 0:35
  Thank you to Britta for supporting the pod, helping us
• 0:38
  keep our deep dives going and of course, keeping us hydrated.
• 0:42
  On September 24th, 2018, a new Instagram influencer came onto
• 0:46
  the scene and his content would become far more influential than
• 0:51
  probably even he could imagine. Standing outside a landmark in
• 0:55
  Scotland, the influencer posted his first selfie with the
• 0:58
  caption Quote the beautiful Dune Castle, also film location for
• 1:03
  Monty Python hashtag John Pork. If you're familiar with the
• 1:08
  iconic John Pork, then you'll know he's not a real human man.
• 1:12
  John Pork is an AI generated virtual influencer, depicted as
• 1:17
  a man's body with a pig head. And though he's certainly not
• 1:20
  the first virtual influencer to crop up on social media, John
• 1:24
  Pork and the various memes surrounding his likeness became
• 1:28
  one of the most notorious, penetrating the mainstream and
• 1:31
  introducing millions of social media users to an early
• 1:35
  iteration of AI slop. This is an exploration into the
• 1:40
  rise and consequences of AI. I slob how and why the Internet
• 1:44
  has become overrun with disturbing and bizarre mass
• 1:48
  produced brain rock content devoid of meaning, and if that
• 1:51
  means the Internet actually is becoming less human or maybe
• 1:56
  even dead. So if you've ever watched a
• 1:58
  TikTok and thought who made this and why does it have 10 million
• 2:02
  views? This one's for you from
• 2:05
  Centennial World. I'm Lauren Meisner and this is
• 2:08
  infinite scroll, a podcast about Internet culture, the creator
• 2:12
  economy and how social media is shaping Gen.
• 2:15
  Z Welcome back to infinite scroll, everyone.
• 2:19
  I'm excited for this episode. I actually can't believe we
• 2:21
  haven't talked about AI slop yet.
• 2:23
  I know we mention it a lot like in this podcast and also on
• 2:26
  mini, and I think Annabelle mentions it a lot on mini tech
• 2:29
  just in context of different stories, but diving into the
• 2:33
  phenomenon of AI slob. I can't believe we haven't done
• 2:36
  this before. So before we get into it, as
• 2:38
  always, Please remember to rate, review and subscribe.
• 2:41
  We're trying to get to 1000 Spotify star ratings.
• 2:43
  Please leave us an Apple review as well.
• 2:45
  If you listen on Apple. I still read the reviews every
• 2:48
  single day and it just like makes me so happy.
• 2:51
  When we get new reviews, please sign up to our newsletter, which
• 2:54
  is also called Infinite Scroll. On Fridays, we round up the five
• 2:57
  biggest thing pieces from the Internet that week.
• 3:04
  The AI boom has defined the past three years or so online.
• 3:08
  Whether through the widespread use and accessibility of ChatGPT
• 3:11
  or various AI powered image generating tools, it's no secret
• 3:15
  that we have entered a new era of online culture.
• 3:18
  And with that has come a lot of discourse.
• 3:21
  Many pundits have expressed their concern over AI, pointing
• 3:25
  to its ability to spread misinformation, take jobs from
• 3:28
  journalists, and as a result, rock the media landscape as we
• 3:32
  know it. However, one of the arguably
• 3:35
  unforeseen consequences of the AI boom has been the rise of AI
• 3:39
  generated slop across the Internet.
• 3:42
  Slop typically refers to nonsensical AI generated
• 3:46
  content. This may be AI generated text
• 3:49
  that is completely full of mistakes or is written in an
• 3:52
  unnatural way, or it may be images that have that weird AI
• 3:55
  generated smooth look to them on content that makes no sense.
• 4:00
  According to reporting from the New York Times Times, the use of
• 4:03
  slop as a descriptor originated in 2022.
• 4:06
  Many have gone on to attribute a developer named Simon Wilson as
• 4:10
  one of the early adopters of the term, but Wilson told the
• 4:13
  journalist quote I think I might actually have been quite late to
• 4:16
  the party. In the piece, journalist
• 4:19
  Benjamin Hoffman explains quote The term sprung up from 4 Chan
• 4:22
  Hacker News and YouTube comments where anonymous poster sometimes
• 4:26
  project their proficiency in complex subject matter by using
• 4:30
  in Group language. It isn't necessarily really
• 4:32
  surprising that AI slop and the word itself originated on select
• 4:37
  platforms and within niche communities.
• 4:39
  This is often the case when it comes to memes and other online
• 4:42
  phenomena before they hit the mainstream.
• 4:45
  But one of the earliest examples of AI slop that entered the
• 4:48
  online zeitgeist in a meaningful way was the AI generated image
• 4:52
  of Pope Francis wearing a puffer jacket from early 2023.
• 4:57
  The picture was initially shared on Facebook and Reddit, where
• 5:00
  the original poster was subsequently banned from Reddit
• 5:02
  hours after posting. Many Internet users went on to
• 5:05
  believe that the image was legitimate, using it to quote
• 5:08
  criticize the Catholic Church for lavish spending.
• 5:12
  To be honest, this incident was definitely one of the least
• 5:15
  harmful and weird instances of AI slob.
• 5:18
  But it's also the perfect example of how AI generated
• 5:21
  content is a hotbed for misinformation and confusion,
• 5:25
  especially on platforms like Facebook.
• 5:27
  Many people attribute this to the fact that older generations
• 5:31
  predominantly use Facebook, claiming that people with less
• 5:34
  media literacy and familiarity with the online space may be
• 5:38
  easier to quote UN quote dupe with slop.
• 5:40
  But of course, the vast amount of AI slop on Facebook may
• 5:44
  partly also come down to issues around metas, moderation
• 5:47
  policies, or a user's interest and engagement farming.
• 5:51
  But more on this later. As Charlie Worzel explained for
• 5:54
  The Atlantic last year quote, AI slop is most prevalent on
• 5:58
  platforms such as Facebook, where click farmers and spammers
• 6:01
  create a collaborate networks to flood pages and groups with
• 6:04
  cheap fake images of starving children and Shrimp Jesus in the
• 6:09
  hopes of going viral, getting likes, and picking up creator
• 6:13
  bonuses for online engagement. This kind of content has even
• 6:17
  spilled over into the political sphere, with Rozelle in the same
• 6:20
  piece for The Atlantic characterizing the MAGA
• 6:23
  aesthetic as AI slop. But that honestly deserves its
• 6:26
  own dedicated deep dive. While AI slop can take many
• 6:30
  different forms, it typically shares a few defining
• 6:33
  characteristics. As Ryan Broderick pointed out in
• 6:36
  an edition of the Garbage Day newsletter last year, this kind
• 6:39
  of content often feels worthless, yet is also
• 6:43
  inescapable. He outlined a couple key traits
• 6:46
  of AI Swap. It is mass produced, imposed on
• 6:49
  users rather than sought out, and primarily driven by
• 6:52
  algorithms rather than organic human interest.
• 6:55
  Despite its perceived lack of value, it spreads quickly,
• 6:59
  dominating feeds and shaping online spaces and ways that are
• 7:03
  hard to ignore. Broderick even takes it so far
• 7:05
  as claiming that AI Slop isn't confined to social media and AI
• 7:10
  alone, but rather content more broadly.
• 7:13
  He goes on to argue that Slop is spreading across digital
• 7:16
  platforms, including corporate giants like Netflix.
• 7:19
  Quote 2018 seems to be the part of Netflix's Slop era.
• 7:23
  In December, it released Bird Box, the last time I remember
• 7:27
  pressing play on a Netflix movie expecting it to be good,
• 7:30
  Broderick writes. This isn't to say Netflix
• 7:33
  doesn't make art anymore, but after 2018 the production of
• 7:37
  Slop clearly outpaced everything else and now stumbling across
• 7:41
  something I like on there is a fun surprise rather than the
• 7:44
  defaults. From Brodericks analysis it is
• 7:47
  clear just how frustrating the onslaught of Slop can be.
• 7:50
  Other analysts have made similar claims, and Rebecca Jennings is
• 7:53
  one such example. In a piece published for Vox
• 7:56
  last year, she spoke to the rise of AI slop within the aesthetic
• 8:00
  fall content niche. While we're used to seas of
• 8:03
  orange leaves, pumpkin spice lattes, and Christian girl
• 8:06
  autumn content flooding our feeds during this time of year,
• 8:10
  she made the claim that fall inspo is essentially just AI
• 8:13
  slop these days. She went on to share how this is
• 8:16
  making online spaces more frustrating.
• 8:18
  While it is quite a niche example, she pointed to the
• 8:21
  online knitting community, noting that they were quote
• 8:24
  being inundated with nonsensical AI patterns and inspo images.
• 8:28
  Users, in turn, had to sift through fake content, blurring
• 8:32
  the line between what's real and fake.
• 8:34
  As the slob becomes the new normal quote, the crux of the
• 8:38
  issue now is the sheer scale of it, Jennings continues.
• 8:41
  Scammers and spammers can unleash a barrage of texts and
• 8:45
  images with the click of a button, so searches for
• 8:48
  legitimate information or a casual scroll through social
• 8:51
  media require even more time. Time and effort to bypass the
• 8:55
  junk. The vast amount of this type of
• 8:57
  content has led to the rise of various theories around what
• 9:00
  this means for social media users and the Internet at large.
• 9:04
  The principal theory among them, of course, is the dead Internet
• 9:07
  theory. Now, we have discussed this
• 9:10
  before on the podcast, mostly on mini scroll episodes, and we
• 9:13
  even touched on it in last week's Infinite Scroll, but it's
• 9:17
  honestly such a widespread theory that it warrants further
• 9:19
  analysis here. So let's break it down.
• 9:22
  The dead Internet theory is a conspiracy theory that
• 9:25
  essentially claims the Internet has been taken over by AI.
• 9:28
  In other words, the majority of the content we consume is no
• 9:31
  longer human generated, but instead created by bots and
• 9:35
  various AI tools. The theory has been around for
• 9:37
  years, with many articles and journalists pointing to a post
• 9:40
  titled dead Internet theory. Most of the Internet is fake, as
• 9:44
  one of the earliest accounts of this theory, but it was a 2021
• 9:48
  article by Caitlin Tiffany in The Atlantic that marked one of
• 9:51
  its earliest discussions in stream media.
• 9:54
  As she explains quote, Dead Internet theory suggests that
• 9:57
  the Internet has been almost entirely taken over by
• 10:00
  artificial intelligence. Like lots of other online
• 10:03
  conspiracy theories, the audience for this one is growing
• 10:06
  because of discussion led by a mix of true believers, sarcastic
• 10:10
  trolls, and idly curious lovers of chitchat.
• 10:14
  As Tiffany touches on the dead, Internet theory has been met
• 10:17
  with a lot of pushback. Many analysts and Internet
• 10:20
  culture enthusiasts remain unconvinced by the theory.
• 10:23
  But that's not to say that there aren't more and more people
• 10:26
  acknowledging the sheer amount of AI content taking over our
• 10:29
  feeds. In a recent journal article
• 10:31
  published in AI and Society, author Yoshiya Walter reinforces
• 10:35
  his claim, writing quote 10 years ago.
• 10:38
  The theory used to be rather speculative, but with the wake
• 10:41
  of generative AI, it can now be observed first hand and it
• 10:45
  highlights a disturbing trend, the blurring lines between human
• 10:49
  and AI driven interactions. Instagram Reels is a case in
• 10:52
  point. Reels is flooded with AI
• 10:55
  generated brain rods. While not all of us may be so
• 10:58
  lucky to fall on this side of the algorithm, Reels has
• 11:01
  everything from Minions to John Pork memes to Skippity inspired
• 11:05
  clips of famous cartoons fighting it out.
• 11:08
  In a recent edition of his newsletter, The Nerd, Adam
• 11:11
  Alexic listed some of the most popular content on reels, noting
• 11:15
  quote AI generated content has gone from being the anomaly to
• 11:19
  being the norm. Stock characters like LeBron
• 11:22
  James get 10s of millions of views when edited into absurd
• 11:25
  scenarios like milking a cow or jumping out of an airplane or
• 11:29
  kissing baby Shrek. There is no denying that much of
• 11:32
  this content is funny, is not insane, but it is also slob,
• 11:37
  offering no real value to social media users.
• 11:40
  Alexik and his piece continues by noting how pervasive AI
• 11:44
  generated content is on the platform.
• 11:46
  Scrolling through 20 videos on his feed, only four of these
• 11:49
  clips were not made by AI. Though this data was definitely
• 11:53
  skewed based on previous content he had consumed and engaged
• 11:56
  with, he notes that quote each of those reels had millions of
• 12:00
  engagements from real users, meaning that those numbers
• 12:03
  wouldn't be off base for many normal.
• 12:06
  These claims have been corroborated by other research.
• 12:09
  Pulling from a study published in Cornell University's preprint
• 12:12
  server Arcsif, Fast Company journalist Chris Stockel Walker
• 12:16
  claims that AI slop is quote suffocating the web.
• 12:19
  Analyzing more than 30 million documents ranging from customer
• 12:23
  complaints to press releases to job postings, the researchers
• 12:27
  tracked the growth of AI generated content online before
• 12:30
  and after Chat GPT's release. Quote Following the November
• 12:33
  30th, 2022 release of Chat ChatGPT, the estimated
• 12:37
  proportion of content in each domain that saw suggestions of
• 12:40
  AI generation or involvement skyrocketed from a baseline of
• 12:45
  around 1.5%. In the 11 months prior to the
• 12:48
  release of ChatGPT, the proportion of customer
• 12:51
  complaints that exhibited some sort of AI help increased
• 12:55
  tenfold, he explains. Ultimately, the surge of AI
• 12:59
  generated content has only fueled dead Internet theory for
• 13:02
  some users. But for the skeptics, let's
• 13:05
  suspend in disbelief for a moment.
• 13:07
  If AI truly dominates the online world, it's easy to see why the
• 13:11
  chronically online would spiral into panic.
• 13:14
  AI generated content and bots threaten the integrity of
• 13:17
  digital information and weaken the community driven essence of
• 13:20
  social media. That said, fear over the
• 13:23
  Internet's quote UN quote fakeness isn't new, it's just
• 13:27
  evolving with technology. Writer Robert Mariani pointed to
• 13:30
  some other instances of similar panic for The New Atlantis in
• 13:34
  2023 for context and transparency, This journal was
• 13:38
  founded by a socially conservative think tank in
• 13:40
  Washington, DC. Nonetheless, Mariani equates the
• 13:43
  rise and panic around dead Internet theory to that of
• 13:46
  Gamergate and various quote, Russian troll farms that emerged
• 13:50
  during the 2016 U.S. presidential election.
• 13:53
  Quote Beneath these panics is a collective gut instinct that
• 13:57
  hiding behind the one man, one account facade of the Internet
• 14:01
  could all too easily be something else and impersonal
• 14:04
  manipulative force. This is the substance of the
• 14:07
  dead Internet theory, the sign of the web user who feels the
• 14:10
  weight of garbage that's been choking the web for the last
• 14:14
  several years. Now, this is definitely not a
• 14:16
  perfect comparison, but it does speak to a challenge that has
• 14:20
  come to define the digital age, one that will only be
• 14:23
  intensified by AI generated slop.
• 14:25
  Despite this, platforms are doing almost nothing to limit
• 14:29
  it. If we consider Reels again for a
• 14:30
  moment, many Internet culture commentators and experts have
• 14:34
  taken issue with the way Meta approaches this kind of content,
• 14:37
  using the output of AI generated content to their benefits.
• 14:41
  First off, Instagram champions engagement creators aren't
• 14:44
  necessarily rewarded for quality content, but rather the
• 14:47
  frequency that they post and of course the amount of users who
• 14:51
  interact with their posts. As Rebecca Jennings perfectly
• 14:55
  puts it for Vox quote, this moment of peak slop is the
• 14:58
  natural culmination of platforms that incentivize virality and
• 15:02
  engagement at all costs, no matter how low quality the
• 15:06
  content happens to be. As for how AI plays into this,
• 15:10
  the pace at which creators can push out AI generated content is
• 15:14
  much quicker than UGC and other types of videos.
• 15:17
  And in a world where engagement is king, it makes sense why
• 15:20
  creators would start prioritizing these kinds of
• 15:22
  videos. On top of this, Meta has fully
• 15:25
  embraced AI, making it clear that they are unbothered by the
• 15:28
  way this content is taking over the Internet.
• 15:31
  For instance, the company previously introduced a suite of
• 15:34
  AI generated characters. While they were later removed,
• 15:37
  Meta went on to launch an AI studio, a way for users to
• 15:40
  create AI characters and chat bots.
• 15:43
  More recently though, Meta has started testing AI generated
• 15:46
  comments. As reported by TechCrunch, Meta
• 15:49
  AI will analyze the photo before generating 3 suggestions for
• 15:52
  comments. Another important factor to note
• 15:55
  about the AI slop on Instagram Reels is that it can be very
• 15:58
  unpredictable. Users deeply entrenched in brain
• 16:00
  rot and all the lore around it would have a high literacy and
• 16:04
  AI slop, but for those who aren't, it may often catch them
• 16:07
  off guard. The thing about brain rot and AI
• 16:09
  generated memes is that a lot of it is self referential, meaning
• 16:13
  if you haven't seen the origins of a particular meme, it will
• 16:16
  probably just leave you confused.
• 16:18
  And considering how rogue the content on Instagram reels can
• 16:21
  get, it's easy to feel this way. But this is also a part of the
• 16:26
  AI slops power. In a piece for The Atlantic at
• 16:29
  the end of last year, Caitlin Tiffany explored a new strategy
• 16:32
  behind engagement farming, and that is to capitalize on
• 16:35
  confusion. While she wasn't solely speaking
• 16:38
  about AI powered content, this concept can definitely be
• 16:41
  applied to AI generated brain rot and the weirdness that
• 16:45
  ensues on this side of reels. Tiffany in the article discusses
• 16:49
  how she'd come across different videos where she couldn't quite
• 16:52
  tell whether they were a joke or not.
• 16:54
  As she explains, quote RB Fidelman appears to be the
• 16:56
  point. These videos are short, and like
• 16:59
  all other Instagram Reels, they autoplay on a loop.
• 17:03
  That's how they succeed. The people who produce them care
• 17:06
  only that I take the time to wonder and that the loop keeps
• 17:09
  looping while I do. They aren't meant to make you
• 17:11
  angry or upset. They aren't playing off your
• 17:14
  curiosity. They're just trying to confuse
• 17:16
  you. And they work.
• 17:18
  Like I said, this confusion isn't universal.
• 17:21
  It's more likely to affect people who randomly stumble onto
• 17:24
  the side of the Internet, but even those deep in AI generated
• 17:28
  memes and brain rot aren't immune.
• 17:30
  They might instantly recognize some AI content, but with
• 17:33
  certain images or videos they might also find themselves re
• 17:37
  watching to figure out if it's real or fake.
• 17:39
  Let's be honest, we've all been fooled by an AI generated photo
• 17:43
  or clip at least once. In a similar vein, there is also
• 17:46
  rage bait adjacent AI slop that drives engagement.
• 17:50
  Not all of it is designed to make people mad necessarily, but
• 17:53
  much of this content is problematic.
• 17:55
  Scroll through Instagram Reels long enough and you'll
• 17:58
  inevitably run into a stream of racist, homophobic, and
• 18:01
  discriminatory quote UN quote memes.
• 18:04
  It is not worth dissecting the content of these videos
• 18:06
  necessarily, but they often animate wildly racist
• 18:10
  stereotypes or capitalize on painful historical moments.
• 18:14
  And at the end, it's all about creating a spectacle, driving up
• 18:17
  views, and getting people to lock in, for better or worse.
• 18:21
  In a piece for Four O 4 Media, Jason Kobler discusses the goal
• 18:25
  behind these posts, noting that quote, the strategy of these
• 18:28
  types of posts is to make a human linger on them long enough
• 18:31
  to say to themselves what the fuck?
• 18:34
  Or to be so horrified as to comment what the fuck or send it
• 18:37
  to a friend saying what the fuck.
• 18:39
  All of which are signals to the algorithm that it should boost
• 18:42
  this type of content, but are decidedly not signals that the
• 18:46
  average person actually wants to see this type of thing.
• 18:49
  Once again, with social media platforms rewarding engagement,
• 18:53
  they aren't doing much to stop this harmful content spreading,
• 18:56
  even when it veers into problematic territory.
• 18:59
  This has ultimately allowed the creators who are making this
• 19:01
  type of content to monetize discriminatory AI slop.
• 19:05
  In the same piece, Kobler notes that many of these AI first
• 19:09
  creators have managed to make money through this kind of
• 19:12
  content from Instagrams bonus program, which offers direct
• 19:16
  payouts. And with many of these videos
• 19:18
  receiving millions of views, they remain incentivized to post
• 19:22
  because dollars are what comes next.
• 19:25
  Adam Alexic for his newsletter interviewed some of these quote,
• 19:28
  UN quote racist Instagram creators, noting how the pursuit
• 19:31
  of virality is the goal and the, quote, extreme nature of their
• 19:34
  content is merely a by product. The Instagram account AI Network
• 19:39
  TV told Adam that, quote, when a video reaches a certain level of
• 19:42
  grossness or weirdness that I find disturbing, I know it might
• 19:46
  go viral. So I export it and post it.
• 19:49
  With social media now driven by virality and profit, things are
• 19:53
  looking pretty bleak. Chasing engagement isn't new.
• 19:56
  It's been the game for over a decade.
• 19:58
  This is especially true since TikTok took over as the main
• 20:01
  cultural engine. And one major consequence, lofty
• 20:04
  as it sounds, is the erosion of human creativity and sincerity
• 20:09
  and content. Over time, we've seen the power
• 20:12
  of content shift. Take online activism, for
• 20:15
  example. While the online efforts during
• 20:17
  the 2020 Black Lives Matter movement weren't perfect.
• 20:20
  They weren't overrun by AI generated noise drowning out the
• 20:24
  real voices and lived experiences.
• 20:26
  Now content and virality are the priority, not the message users
• 20:31
  or the creator behind it, pushing creativity and
• 20:34
  earnestness to the back burner. As Damon Bears puts it for the
• 20:37
  Atlantic quote, AI SOP is lubricant for broken algorithmic
• 20:41
  machinery poured into a social network.
• 20:44
  And if you've done the alchemy right, the gears will turn and
• 20:47
  turn. This is the Internet synthetic
• 20:49
  maximalist moment where fake content leads easily to
• 20:52
  superficial interaction. Aiden Walker in his Sub Stack
• 20:56
  newsletter How to Do Things with Memes, expanded on what
• 20:59
  replacing human content and communication with AI slop
• 21:03
  actually means. First, he suggests that the
• 21:05
  decline of human generated content ensures the balance of
• 21:09
  power remains with the platform. As Walker explains quote, slop
• 21:13
  helps control people. If through AI slop, they can
• 21:16
  crowd out actual human voices on platforms, not only will they be
• 21:20
  free from the obligation of paying creators, but they will
• 21:24
  control more of the conversation.
• 21:26
  As we've already touched on, there's been growing discourse
• 21:29
  and commentary circles about how users are more preoccupied with
• 21:33
  content itself over the actual creativity.
• 21:37
  We've seen this shift from communication to consumption,
• 21:40
  where users increasingly visit platforms to consume content
• 21:43
  rather than connect with one another.
• 21:45
  This ultimately reflects just how commercialized the space has
• 21:48
  become. While many creators still aim to
• 21:51
  produce meaningful content and encourage community, others
• 21:54
  fully embrace the churn. This speaks to how our content
• 21:57
  consumption habits have shifted. As I mentioned last week, we
• 22:00
  have become incredibly passive consumers with algorithms and
• 22:04
  recommended pages dictating what we see.
• 22:06
  This has made it easier to engage with content for content
• 22:09
  sake. AI slop and low quality posts of
• 22:12
  that nature included. To return to the era of quality
• 22:15
  content, the solution requires collective action from social
• 22:19
  media users. Quote it has to come from us,
• 22:21
  the user, the viewer, the consumer and there's a lot of us
• 22:25
  now we have to be the ones to demand that we all make less,
• 22:28
  aim smaller, be more deliberate about what we consume and find
• 22:32
  new ways of funding and distributing what we do make.
• 22:36
  Ryan Broderick concludes another major consequence of the rise of
• 22:40
  AI slop is that AI generated content is actively limiting
• 22:44
  human to human interaction online.
• 22:46
  Take the recent Studio Ghibli AI trend as an example.
• 22:49
  Users worldwide we're quick to jump on the trend, using ChatGPT
• 22:54
  to giblify various pictures. It kicked off after Open AI
• 22:57
  launched its Images for ChatGPT tool, allowing users to generate
• 23:02
  images within ChatGPT itself. Soon after this feature went
• 23:05
  live, people began using it to transform existing pictures into
• 23:09
  images that looked like Studio Ghibli animation to them.
• 23:12
  The trend sparked concerns around the ethics of AI pulling
• 23:15
  from different artistic styles. Users are concerned about Open a
• 23:19
  Eyes failure to protect artists intellectual property.
• 23:23
  Thinking about the labor that goes into creating these
• 23:25
  animations, it's hard not to see the problem.
• 23:28
  Open AI replicates the hard work of these animators without
• 23:31
  crediting or compensating the original artist.
• 23:34
  While Studio Ghibli definitely has received the most traction,
• 23:37
  this isn't the first time Internet users and artists have
• 23:40
  expressed concern. In 2023, for example, Kyle Chaka
• 23:44
  reported for The New Yorker on a class action lawsuit against MID
• 23:48
  Journey and two other AI imagery generators on behalf of artists.
• 23:52
  Quote. In some sense, you could say
• 23:54
  that artists are losing their monopoly on being artists.
• 23:57
  With generative AI, any user can become an author of sorts.
• 24:00
  So where does that leave us? Well, for now, we're scrolling
• 24:05
  through platforms increasingly saturated with SOP, confused,
• 24:09
  entertained, horrified, and most of all, exhausted.
• 24:13
  The Internet is overrun with algorithmically generated junk
• 24:17
  that doesn't just dominate our feeds, but distorts our
• 24:19
  perception of what's real, what's valuable, and what's even
• 24:23
  worth engaging with. And as we're drowning out human
• 24:25
  presence, online platforms don't seem to care as long as we're
• 24:29
  watching our feeds feel less like reflections of human
• 24:33
  experience and more like simulations of it.
• 24:36
  Content made for us, but not by us.
• 24:39
  And that's possibly the scariest part.
• 24:42
  If AI slop continues to replace creativity and communication,
• 24:46
  what does that mean for the future of the Internet?
• 24:48
  And if real people are being drowned out by content farms,
• 24:51
  fake engagement, and algorithmic manipulation, where does that
• 24:55
  leave the social in social media?
• 24:57
  That's a deep dive for another day.
• 25:00
  Infinite Scroll is a podcast produced, presented and Co
• 25:03
  written by me Lauren Meisner for Centennial World.
• 25:06
  This episode was researched and Co written by Annabelle and
• 25:09
  David. Thank you so much for listening.
• 25:11
  I love you guys and I will talk to you next week.
• 25:14
  Bye.