Latest / Mini Scroll / MINI TECH SCROLL: Chappell Roan bot backlash, response to the social media addiction lawsuits + is KATSEYE using AI?

Transcript

MINI TECH SCROLL: Chappell Roan bot backlash, response to the social media addiction lawsuits + is KATSEYE using AI? · Mini Scroll

0:00
Hey everyone, and welcome to Mini Scroll, the daily Internet
0:03
culture podcast bring you the biggest stories from social
0:06
media, the creator economy, and the digital space every Monday
0:09
through Thursday and sometimes Fridays.
0:11
My name is Annabelle and I am your Wednesday Mini Tech Scroll
0:14
host and also the deputy editor of the Australian youth digital
0:18
media brand Centennial World. Welcome back to Mini Tech Scroll
0:22
for another week. I hope you guys are doing well.
0:25
Before we get into this week's episode, I just wanted to
0:28
acknowledge the lighting change. I feel like the lighting looks
0:32
so different right now because of the time of year we're in.
0:34
And obviously we have a window now.
0:36
So it's like a completely different vibe.
0:38
And while it is definitely a fun change, I do want to say if you
0:42
ever see me squinting, it's because the sun is like
0:44
progressively moving as I'm recording and it like kind of
0:48
gets in my eye. So that's a bit of fun.
0:50
But I just needed to acknowledge that I don't have it too much to
0:54
say at the top of this week's episode apart from our classic
0:57
reminder to rate, review and subscribe.
1:00
Subscribe and also feel free to check out our Patreon if you
1:03
want more behind the scenes content of Centennial world.
1:06
OK, let's get into the content for today.
1:09
And for our first story today, Lauren actually teased that I
1:13
would be covering this on Monday.
1:15
And that is of course an update on the Chapel roan security
1:18
guard situation because recent reporting suggests that the
1:21
strong anti Chapel sentiment may be a result of a coordinated bot
1:25
campaign. I genuinely think Lauren covered
1:28
this story like four to five times last week and that is why
1:32
we decided to cover this part of the story in mini tech just to
1:36
give a little bit of a reprieve. But I will give a very quick
1:40
summary of everything that went down before we get into the bot
1:43
component. So this all started like the
1:45
weekend before last when Chapel Roan found herself at the center
1:49
of controversy after Brazilian soccer player Georginio called
1:53
out the singer on his Instagram Stories.
1:56
He alleged that Chapel sent a security guard to talk to his
1:59
wife instead stepdaughter after his 11 year old stepdaughter
2:02
walked past Chapel's table and allegedly just looked at her,
2:06
and this quickly snowballed. Many social media users were
2:09
disappointed to hear about Chapel's alleged attitude as
2:13
well as the behavior of her security team.
2:15
After this initial wave of backlash, Chapel posted a video
2:18
to her Instagram story clarifying that the security
2:21
guard was not part of her personal security detail.
2:24
The security guard later corroborated her story saying
2:27
that his actions were not on behalf of her or her personal
2:30
security team or her management. As you probably saw, there was
2:34
so much hate directed towards Chapel after this and it was
2:38
arguably out of proportion to the offence.
2:42
Granted, there is a constant conversation around Chapel and
2:45
her strict boundaries around fan interaction, so this quote UN
2:49
quote hate did kind of fit neatly into that discourse.
2:53
On top of this, Chapel is a pretty polarizing figure whether
2:58
she intends to be or not. For instance, her political
3:01
positions are routinely criticized by both the left and
3:05
the right. She is also an openly queer
3:07
artist in an online environment that is increasingly
3:10
conservative. And also like I've already
3:12
mentioned, she is very willing to set boundaries, which is not
3:16
something many people like to see or are at least used to from
3:20
women in a public way like this. So with all of this and the
3:24
situation at hand, there were quickly thousands of posts
3:28
talking about how Chapel is evil.
3:30
She's entitled with memes and commentary circulating about how
3:34
she hates children. And funnily enough, this Chapel
3:37
hates children discourse has actually entered Internet
3:41
discourse before, particularly after Chapel spoke about how her
3:45
friends with kids are in quote UN quote hell.
3:48
All of my. Friends who have kids are in
3:50
hell. I actually, I don't know anyone
3:51
who's like happy and has children at this age.
3:54
Anyone who has like light in their eyes, anyone who has, he
3:56
was. Slept and the response to this
3:58
situation may have come off the back of that, but it also may
4:03
have been because Georgino is well loved and the Internet
4:07
wanted to side with the soccer Stein his child or perhaps
4:10
because people just kind of love to hate on Chapel or they feel
4:14
as if she did something wrong in this scenario.
4:16
According to recent reporting by BuzzFeed, The reason behind this
4:20
level of backlash and the intensity of the backlash with a
4:24
little bit more sinister. The publication spoke with
4:26
Gudea, which is an AI powered platform that identifies the
4:30
quote origins of online information, how it will move
4:33
across platforms, and where it will have the most impact,
4:36
according to its little LinkedIn description.
4:39
And according to Gudea, the negative response to Chapel's
4:42
situation was driven by bots. Gudea reportedly examined around
4:47
100K posts by approximately 54 K unique uses across various
4:52
platforms between March 20 and March 20 two, and they alleged
4:57
that only 6.5% of the posts they reviewed were original content
5:01
where the rest were reposts. As Gudea explained on LinkedIn
5:06
quote, 4.2% of total users showed signs of inauthentic
5:10
behavior. That number is exceptionally
5:12
high and it wasn't evenly distributed.
5:15
The narratives attracting the most inauthentic activity were
5:18
the most culturally charged, particularly those targeting her
5:21
LGBTQ plus identity. This is what coordinated
5:24
narrative campaign pains look like in practice.
5:27
For some additional context on Gudea, they have kind of done
5:31
some similar investigations in the past, most notably around
5:35
Taylor Swift and Bad Bunny. I will say I have seen some
5:38
users in the past kind of criticize Gudea and their
5:41
research methodology, saying that it isn't really clear
5:44
enough and that obviously we can't necessarily take the
5:47
findings from one report and run with it.
5:50
And I think that is definitely a fairpoint, but if we look at the
5:54
way the scandal evolved with Chapel, it kind of fits into how
5:58
a smear campaign might look like.
5:59
The conversation first started with the situation in the hotel
6:03
with the security guard, but it quickly turned and devolved into
6:07
other topics around Chapel, with many positioning her as a liar
6:10
and a hypocrite. For example, Post started
6:13
circulating that Chapel was lying about her background as
6:15
well as being a lesbian and some Chapel stand accounts.
6:19
Of course these accounts are biased, they do have a narrative
6:23
they want to push forward around their fave.
6:25
But they started pointing out various accounts they believed
6:29
to be bots. As one user on X wrote quote,
6:33
this is an example of a bot being used by whoever is
6:36
enacting and funding this. It never posted about Chapel
6:39
until 2 days ago and is now posting about her non-stop using
6:43
the same videos as other bots and paid accounts.
6:46
So I did look into the account that this user mentioned and
6:49
from what I could see the 1st post about Chapel was from March
6:53
22 and I counted at least 25 between then and March 30.
6:58
Some of these post word duplicates and just copy pasted
7:01
and others were AI slot. Obviously we can't say for sure
7:05
if this was a bot, but I think we can say that a normal account
7:09
or like a normal typical person wouldn't necessarily operate
7:13
like this online. And if it wasn't a bot, it could
7:16
be an account farming for engagement and virality, which
7:19
also could be pursuing a certain narrative.
7:21
Whether you believe the backlash to Chapel was a bot campaign or
7:25
not, we do know that these types of campaigns are becoming
7:29
increasingly prominent online. And I have covered it a couple
7:32
times on mini tech before. And we see them often as large
7:36
scale attempts to alter public opinion or push disinformation
7:41
to change behaviours and attitudes among a specific
7:44
demographic. And it has been used by
7:46
governments, but it is also increasingly being used by non
7:49
state actors in recent years. But probably the most infamous
7:54
example of these coordinated influence campaigns was the
7:57
reported effort to interfere in the 2016 U.S. presidential
8:01
election. Having said that, this raises a
8:04
key question. Why would there be a bot
8:06
campaign targeting Chapel Rd. in the 1st place?
8:09
Now I will note that this is very much a theory and
8:12
speculation, but some social media users are convinced that
8:15
this is coming from from the top down, especially because Chapel
8:18
made headlines after leaving Wasserman Agency after Casey
8:22
Wasserman's name appeared in the Epstein files.
8:25
Now you can obviously believe what you want about Chapel Roan
8:29
and whether this is a smear campaign or not.
8:31
I know she's a super polarizing figure and people are not
8:35
necessarily willing to give her the quote UN quote benefit of
8:38
the doubt, but I think that is also a very important point
8:42
here. It can already be hard to clock
8:46
when bots are posting online, and it seems especially
8:49
difficult when it comes to pop culture because of how
8:52
inflammatory this space is. Specifically where these one
8:56
sided posts easily fit into the way we communicate the online
9:00
zeitgeist and just the digital landscape at large.
9:03
And when women, not exclusively Chapel, are categorized as
9:07
difficult or entitled, butts that align with this messaging
9:11
can operate as a form of confirmation bias for users,
9:15
allowing this message to stick even further in some respects.
9:19
I do think the certain descriptors that are used to
9:22
describe some women, like difficult, can be an attempt to
9:26
justify misogyny by labeling her as someone who kind of deserves
9:30
this vitriol. And hatred, not only does this
9:34
end up affecting a woman's career, but men are rarely
9:37
targeted in the same way and receive this specific criticism.
9:41
So once again, we really see the patriarchy online working
9:45
overtime. OK, let's move on to the next
9:48
story. I wanted to cover all the
9:50
discussion that has emerged around the social media
9:52
addiction lawsuit in the US. Now I wasn't too sure if I was
9:56
going to cover this because sometimes I feel like we do get
9:58
a little bit US centric on mini tech scroll.
10:01
But this conversation has just been dominating my social media
10:04
algorithm and feeds over the past week or so, so I thought it
10:07
could be relevant to some of you guys if you've been seeing it as
10:10
well. I'll first start off by giving
10:12
you a little bit of context around the lawsuits.
10:14
We're obviously not a legal podcast, so I will give some
10:18
background, but predominantly I want to focus on the online
10:21
discourse around it. Essentially, a jury in LA found
10:25
Meta and YouTube liable for deliberately designing addictive
10:28
products. After what was like a six week
10:30
long trial, the case was brought forward by a now 20 year old
10:33
woman named Kaylee, and it was actually first filed in 2023
10:37
when she was 17 years old. It's being a widely reported on
10:41
and considered as a kind of quote UN quote bellwether trial
10:45
of sorts to kind of see how the jury and the judge would react.
10:48
So Kaylee's story was that she had created YouTube, Instagram,
10:52
Musically, and Snapchat accounts all between the ages of 6:00 and
10:55
11:00. She alleged that she had become
10:58
addicted to these platforms as a child because of the features
11:01
designed to promote engagement, like infinite scroll and like
11:05
autoplay. She claims she subsequently
11:07
experienced anxiety, depression, and body image issues because of
11:11
it, and her addiction to social media.
11:13
Meta and YouTube obviously pushed back on this, pointing
11:16
out the fact that she had a troubled background and home
11:19
life where they could be contributing factors to her
11:22
mental health struggles. I should also add that Snapchat
11:26
and TikTok were also involved in this case but actually settled
11:30
out of court and this news broke and I think it was January.
11:34
But what was significant about this trial was that it treated
11:37
platforms like YouTube and Instagram as defective
11:40
themselves. The focus in this trial wasn't
11:43
whether there were bad actors posting problematic content, but
11:47
rather that the platforms themselves and the features they
11:49
introduced uphold it and are therefore dangerous.
11:53
I should add that social media platforms are generally legally
11:56
protected from being liable for the content posted on their
12:00
platforms in the United States. And I know I'm about to get a
12:03
little bit granular here, but it is relevant.
12:05
So they are protected from this because of Section 230 and the
12:09
Communications Decency Act. And it essentially says that if
12:12
a user posts something on platform, the platform can't be
12:16
sued for that. It is on the person who posted
12:19
it. So this case kind of went around
12:21
that by focusing on design, and it seemed to have worked.
12:25
A couple days ago, the jury handed a verdict in favor of
12:28
Kaylee, the plaintiff against Meta and Google.
12:31
The tech giants now have to pay her 6,000,000 USD in damages,
12:35
with Meta to pay 70% and YouTube to pay the remainder.
12:39
Meta and Google are set to appeal this result.
12:42
YouTube continues to say that it isn't really a social media
12:45
platform and shouldn't be treated like that.
12:47
This is something they did in Australia with the under 16
12:50
social media ban. But aside from this inevitable
12:53
appeal, there are also thousands of similar cases that are set to
12:56
go forward across various courts in the United States over the
12:59
next year or so. While Kaylee's lawsuit was one
13:02
factor, there was another lawsuit that met a loss that
13:05
week, and it was in New Mexico. It was brought by the state
13:09
attorney general that essentially claimed that design
13:11
and choices had facilitated the sexual exploitation of children,
13:15
and Mehta were ultimately told to pay 375 million in damages.
13:19
So it was definitely a bad week for Mehta, even though this
13:22
money is really just a drop in the pond for them.
13:24
OK, Now that we have the context, I wanted to talk a
13:27
little bit about the response online to Kaylee's judgment
13:31
specifically. And I would also love for you
13:33
guys to share your thoughts. So there seems to be two main
13:37
sides or camps reacting to the verdict.
13:41
There's been a lot of coverage and discourse saying that this
13:44
is a major win for consumers and those of us who are against Big
13:48
Tech. There's been a lot of comparison
13:50
to previous cases against Big Tobacco in the 1990s and the
13:54
early 2000s as well, that essentially found that tobacco
13:57
companies were conspiring to conceal the health risks of
14:01
smoking and also targeted youth and all these other things that
14:04
were very negative for the industry at large.
14:08
So I wanted to read a little bit from an opinion piece from The
14:11
New York Times by investigative journalist Julia Ann Gwynn, who
14:15
praises the verdict and kind of epitomizes this sentiment as she
14:20
writes. Quote.
14:21
With this ruling, the new era of holding tech companies
14:23
accountable for their antisocial choices has finally begun.
14:27
The truth is it's hard for profit motivated companies to
14:30
decide to make less money so we can raise the cost of bad
14:33
behavior. It worked with smoking.
14:35
You can still smoke, but it could cost $15 a pack and you
14:39
often can't find a place where it's allowed these days.
14:41
Similarly, there's no need to ban social media, we just need
14:45
it to be too expensive for these companies to keep doing business
14:48
as usual. I've also seen a lot of comments
14:51
from social media users who feel kind of vindicated with this
14:55
verdict. And I thought I would read a few
14:57
comments from the AUS slash technology Reddit that kind of
15:00
show you this perspective. As one user wrote quote.
15:04
As a strong survivor of social media, the verdict encourages
15:07
me. It shows me that I can make my
15:08
own case in the future. And another wrote quote, These
15:12
algorithms are so hidden and we need transparency.
15:14
These companies referring to big tech are the new big tobacco of
15:18
our age. It should be regulated same way
15:21
as those or even more so. That is one side of the
15:24
response. And I will say I do think that
15:26
is a lot of people's immediate reaction upon hearing this
15:29
verdict. On the flip side, though, I have
15:31
seen various tech generations, analysts and social media users
15:34
express more caution around the verdict.
15:37
They suggest that with this verdict, there is a slippery
15:40
slope to a completely surveilled Internet.
15:42
With a judgment like this one, there is clearly increased
15:46
political incentive to get children off these platforms to
15:49
avoid causing them harm. And this in turn increases
15:54
pressure on tech platforms to moderate.
15:56
And that might require things like age gating, identity
15:59
verification, and just increase censorship.
16:02
And a lot of people argue here that this loss of anonymity
16:05
online will put various marginalized groups in danger
16:09
and restrict free speech in the process.
16:11
There are also questions whether this kind of moderation will do
16:15
anything for social media addiction.
16:16
In the fast place, the CEO and founder of the blog Tech dot,
16:20
Mike Masnick question this in a piece and I'll read it out to
16:24
you because I feel like he frames it quite well.
16:26
Quote Here's a thought experiment.
16:28
Imagine Instagram, but every single post is a video of paint
16:31
drying. Same infinite scroll, same
16:33
autoplay, same algorithmic recommendations, same
16:36
notification systems. Is anyone addicted?
16:39
Is anyone harmed? Is anyone suing?
16:41
Of course not. The addictive design does
16:44
nothing without the underlying user generated content that
16:47
makes people want to keep scrolling.
16:48
I've also seen social media users and journalists on this
16:52
quote UN quote side of the discourse categorize the
16:55
addictive nature of social media as a moral panic, arguing that
17:00
there are insufficient causal claims of social media addiction
17:03
causing mental ill health. I have also seen quite a bit of
17:07
commentary as well around how the comparison between tobacco
17:11
and social media is a false equivalence.
17:13
This was put pretty succinctly in an opinion piece for the New
17:16
York Times by columnist and political commentator David
17:19
French, and I will read that also out to you quote, a social
17:23
media site isn't a bottle of alcohol or a cigarette.
17:26
It's not delivering a drug. It's delivering speech.
17:29
Sometimes that speech is silly and harmless.
17:32
Sometimes it's toxic and harmful.
17:33
Sometimes it's educational or inspiring.
17:36
Aside from the journalist I've already mentioned, Taylor Renz
17:39
has been pretty vocal about this, as you'll see on her
17:41
social media profiles. I've also seen Ryan Broderick
17:45
speak about this topic in the garbage day newsletter titled
17:48
Meta Bad. But losing section 230 was, and
17:52
there isn't a a part of the conversation here that I haven't
17:54
really brought up yet in response to Kaylee's case.
17:57
And that is about the role of parents in monitoring their
18:01
child social media usage. I'm not a parent, so I feel like
18:05
I can't really comment on that. I also completely had unfettered
18:08
access to social media while I was growing up, for better or
18:12
for worse. But I also think the social
18:14
media landscape has changed significantly since like the 20
18:18
tens, which was my point of reference.
18:20
So as you can probably gather, there is so much to this
18:23
conversation. So please think about what I
18:26
have just presented to you as a primer.
18:29
It could literally be a whole deep dive.
18:31
But I'm definitely curious to hear your thoughts on any part
18:34
of the conversation. I will say it's obviously quite
18:36
complicated, especially if you have limited context of the
18:39
United States. But for me here in Australia, it
18:42
is really interesting to hear these perspectives because in
18:45
some ways Australia is quote UN quote ahead, for lack of a
18:48
better word, in that the under 16 social media ban has already
18:53
been implemented along with various age gating measures,
18:56
especially around porn online. Arguably, A verdict like the one
19:00
in Kaylee's case legitimizes these actions, for better or for
19:03
worse. As far as what comes next for
19:05
the US, I'm sure we will continue to see appeals with
19:09
these cases. So I don't think this is going
19:11
out of the news cycle anytime soon.
19:13
OK, We seriously need a lighter story after those two and any
19:17
excuse to talk about Cat's Eye I will take.
19:20
So this brings us to our third story for today that Cat's Eye
19:24
is facing allegations of using AI amid the roll out of their
19:27
new era. So Cat's Eye has been posting
19:30
teasers for their new single which is called Pinky Up.
19:33
The song and choreography has already been leaked, but I have
19:36
not seen favorable reactions towards it.
19:40
And also all the questions about whether or not will be
19:42
returning. But that is honestly a
19:44
conversation for another time. One of the teasers that has been
19:47
posted is the Mona Lisa holding a teacup, with her pinky up of
19:51
course, and then taking a sip. Another one, which was more
19:55
recently released, shows a claw machine picking up a sword.
19:58
Both sound fairly harmless, but social media users have come to
20:02
believe that they were made with AI.
20:04
Some of the top comments on the Mona Lisa teaser.
20:08
Include Quote hype, stop using AI challenge and another one was
20:12
Quote Hype's obsession with AI is actually killing me.
20:15
Those two comments at the time I was researching this had around
20:18
20 K likes each, so I feel like that is definitely a common
20:22
sentiment. Having said that though, it does
20:24
seem like these Internet users may have jumped the gun.
20:27
Many fans of Cat's Eye have noted in response that the
20:30
videos aren't AI, but CGI, which would be the manual creation of
20:34
these visuals. Users are pretty sure that they
20:38
were created by Blunt Action, which is a creative studio based
20:41
in New York City. They've worked with Cats I
20:44
before, specifically in the lead up to Nali, and it does seem
20:49
like in this case, blunt action was involved because they did
20:52
share the teaser to their own Instagram story, which is I
20:56
think actually still live as I'm recording this.
20:58
While the use of AI is a little bit blurry here, there have been
21:01
a number of incidents that have gone viral over the years where
21:04
hype has reportedly used AI. In 2023, it was reported that
21:09
Hive backed artist Midnat used AI to release a track in six
21:13
different languages where the AI was used to assist in the
21:16
pronunciation, tone, and delivery across the different
21:20
languages. Hive also has a majority stake
21:22
in an AI sound startup called Super Tone.
21:25
That isn't even to mention the previous allegations against
21:28
Hive that Cat's Eye uses AI in visuals and even in songwriting.
21:32
When it comes to visuals specifically, there was a lot of
21:36
discussion upon the release of Internet Girl because 351 Studio
21:41
was credited and according to the website, 351 Studio is quote
21:45
a leading AI video production agency delivering cinematic
21:48
generative AI video content and post production services.
21:52
But aside from Cat's Eye, I've obviously covered the alleged
21:55
use of AI and teasers in mini tech before, particularly
21:58
related to Taylor Swift and the roll out of the life of a
22:01
showgirl. But the reason I wanted to
22:03
discuss this situation in relation to Cat's Eye is not
22:06
just because I am obsessed with them, but because of this piece
22:10
that I thought was really interesting that was published
22:12
by the Rolling Stone recently. It was written by Nathan
22:16
Brackett and basically said that the current attitude towards AI
22:19
in the music industry is quote UN quote, don't ask, don't tell.
22:23
Speaking to industry experts and insiders, he finds that AI
22:27
powered tools are being used but only behind closed doors where
22:32
there continues to be kind of a social penalty for openly using
22:36
it, which honestly is not very surprising.
22:39
He says that the industry is essentially relying on the
22:42
honest system at this time. As for whether artists are using
22:45
AI or not, from his conversations though, he says in
22:48
private, many people in the industry are willing to admit
22:51
the merits of AI in the way that it can empower songwriters.
22:55
It can almost like fill a skill gap where they can send around
22:59
songs they write without splitting the profits with a
23:02
producer, let's say. But speaking to the impacts of
23:05
AI more broadly, he writes quote.
23:07
Of course, for every task that AI streamlines, there might be
23:10
someone on the other end who isn't paid anymore.
23:13
A demo musician or producer and assistant engineer who helps
23:16
with mixes. A studio owner renting time,
23:19
maybe a 70 songwriter living off licensing fees.
23:22
I'll link this piece as well in the description because there
23:26
are a lot of interesting quotes from different people about how
23:28
the industry is operating. So if you're interested in that,
23:31
definitely check it out. But back to the hype of it all.
23:35
Obviously the Cat's Eye example is a little bit different to
23:37
this, but I do think seeing a conglomerate like Hi embracing
23:41
AI in some capacity is never not going to cause some level of
23:44
panic and disappointment among fans.
23:47
But I also think the response to Pinky Up specifically shows us
23:51
that we are risking a raising the people and the small
23:54
businesses slash agencies who do the work when it comes to
23:58
creating visuals in this case. But yes, if anyone wants to talk
24:02
about anything Cat's Eye related, I fear I'm locked in
24:05
and always ready to discuss. Well that is everything from
24:09
Mini Tech Scroll today. I really feel like I have been
24:12
recording for a very long time so I'm a bit fearful to edit
24:15
this, but I hope you guys enjoyed.
24:17
Lauren will speak to you guys tomorrow and I will speak to you
24:21
next week. Thanks for listening.
24:23
Bye. Centennial World Podcast Network
24:25
operates on Gadigal land. We would like to acknowledge the
24:28
Gadigal of the Aura Nation, the traditional custodians of this
24:31
land, and pay our respects to elders both past and present.