> what is the actual good-faith rationale for using this feature?
There is none, because the whole thing was never good faith. It's always been about tricking people into having a real interaction, to make it easier to exploit them for ad clicks or to sell them junk. That's the whole YouTube content creator business, and it's what it always has been.
The good faith is "We here at youtube adamantly REFUSE to spend any actual manpower on things that require it, and you know what, so should you! Join us. Thrive. Beep Boop."
Of course they only do it because it's inevitable, and they don't want others doing it before them.
Brand management people, who normally handled this, cost money. Self-service brand management with LLMs costs less money. Building this in as first-party product commoditizes this, costing the vlogger even less, while YouTube gets to keep them in-house and focused on the platform, instead of external services that may be cross-platform.
Yes, so if we assume that cost saving is the only metric that is meaningful and that there should be no legislation or norms saying otherwise, then indeed it follows as a consequence.
Cost saving is basically what drives the market, and thus shapes our entire reality. It's not some side goal of some people, that you could suppress or eliminate - it's the thing itself.
As such, I'm not sure where you'd like to draw the line. There's no obvious, defensible place to do it, not unless you're willing to go back up, expand the scope and fight the entire advertising industry on principle - in which case, I'll agree, because I believe that's where the problem originates and where it needs to be solved.
It is not inevitable, almost nothing is... This just bothers me because so many people in tech talk about things as being "inevitable" when it's just a lazy resignation to the current zeitgeist.
There are a plethora of other forces at work beyond "the market".
Yes, if no one imagines anything different the future will turn out as we expect. But with even small changes to legislation, norms and culture thing can turn out completely different. It is only inevitable if we resign ourselves to the status quo.
> But with even small changes to legislation, norms and culture thing can turn out completely different.
Those are the qualities that define "the market". What are the other forces you speak of?
Markets can change. They have many times before. But, and call me unimaginative if you will, I struggle to see why anyone would want to pay substantially more to ensure that comments on YouTube are written by a real, live human, let alone enough people to sustain the service. It is not like you are face-to-face at the bar. It is a blob of text that has always been disassociated from what human involvement there may have been behind the scenes.
Inevitable in the strictest literal sense may be too much, but the chances of the market changing here seem infinitesimally small to the point that "inevitable" is close enough.
I see this attitude over and over again, particularly where it comes to regulating things like AI and bans on social media. Tech would rather do nothing if "it's complicated", or had any downside to anyone while ignoring the rampant downsides impacting everyone right now. Sometimes it comes across as thoughtful policy making, but more and more I see it as a crutch for intellectual laziness and in some cases dishonesty.
In these cases for it to be inevitable there doesn't need to be agreement. Someone does it and if enough clueless people use it, now we're stuck with it. Why we can't have nice things.
Not just Youtube. Most of the time unintentionally but, any personified information broker is manipulative in its nature. Watching vlogs or streamers, listening podcasts, reading post of social media accounts one follows. There is face between you and the information and that face becomes familiar to you. Face that builds trust.
It used to be that TV or newspapers or any other media was 'talking' to the audience (plural) now it talks to 'you' (singular).
> Most of the time unintentionally but, any personified information broker is manipulative in its nature.
I'm past calling this unintentional. It's not. If it's some random account barely anyone has heard of, posting interesting stuff every now and then, then sure, that's unintentional - and also mostly harmless. However, if the creator is somewhat known[0], posts regularly, and their content is structured and polished, and they start uttering phrases like "like and subscribe" or "ring the bell"[1], or Patreon is mentioned - then you're looking at entirely intentional manipulation.
To a degree, it's unavoidable - it's the nature of the medium and the economy at large. Making quality videos is pretty much a full-time job, so even the creators with purest of hearts will be forced to include stuff that puts their videos on SponsorBlock lists. But then there's a difference between those who want you to subscribe to their Patreon and maybe buy some stickers so they can afford treating their hobby as a job, and those whose content is just a vector for feeding you first-party and third-party (sponsor) ads. Most well-known vloggers are, unfortunately, the latter. Also anyone who's called or calls themselves an influencer is - it's literally the definition of that word.
--
[0] - Perhaps you could say they're "effectively a brand" in their niche - but then, that phrase alone should tell you something.
[1] - Unless it's preceded by "climb the steps" - then you're dealing with a whole other barrel of monkeys.
> even the creators with purest of hearts will be forced to include stuff that puts their videos on SponsorBlock lists
I don't agree with this at all. No one is forced into making video cration their only job. Yes, making videos on weekends means less "output" and as a result a smaller following. But so what? It's still a choice. And IMO almost always a bad one if you care about quality - pretty much no person can keep up creative quality for long when working on a schedule and that's before you even get to intentionally degrading the quality for the sake of monetization.
Every person at the very least has the choice of just creating and sharing things rather than trying to build a business. Of course calling them "creators" already shifts the discussion into the worldview that shapes the current internet.
If "Youtuber" isn't a profitable job without engaging in slimy practices then maybe it shouldn't be a job at all. In fact, trying to make any human activity into a profitable business is one of the big if not the main driving forces behind the enshittification of the internet as well as many things outside it. And it's always this same justification - that engaging in shitty behavior is required to compete.
Personally I don't mind you mentioning your Patreon [0] but "like and subscribe" guarantees a dislike from me and I sure as hell am not going to subscribe [1]. Sponsored sections will get me to immediately close the video and make a mental note to ignore videos from that person in the future. Yes, I could use SponsorBlock but if someone is willing to sell themselves in that way I don't trust them or their videos to not be also compromised in other less obvious ways. It's important to remember that Youtube and similar media is completely optional entertainment and you don't need to engage in any of it.
[0] Please consider alternatives though, Patreon itself is a pretty shit website what I only use begrudgingly.
[1] I don't use youtube subscriptions at all, only RSS subscriptions for the very few channels I want to follow - most things are not important enough to get regular releases from and I'm fine with only seeing videos that are good enough for others to share them with me.
I don't think listening to a podcast is really substantially different to tuning into Walter Cronkite. Both are deliberately recognisable human faces (or voices) for the presentation of the media to you.
Maybe it gets a bit weird when you start messaging your favourite Youtuber and they reply fast enough that you feel like it's a personal conversation rather than a professional correspondence. Sending a letter to Cronkite would have been far less immediate.
Treating media as anything other then virtually entirely read-only has always been what nutters do - they'd be the only ones writing any significant numbers letters to the person if the TV, and the same is probably true of modern media.
No, but the survival of the "entrepreneurs" who convince people they need those rocks, very much is.
In the OF/pet rock/pidgeon trio, the pidgeon is actually the odd one out; the other two are all about making money by exploiting people's need for socializing.
Alas, "e-celebs" are a step worse than "e-prostitutes". Both exploit parasocial relationships to trap customers, but in the end, OF performers provide customers actual value in exchange for money, while vloggers and influencers also try to sell you all kinds of random shit - their actual business is advertising.
(It's similar to analog world prostitutes vs. telemarketers; the former engage in voluntary exchange of value for money, the latter just try to scam you. That the society scorns the former while accepting the latter is some perverse inversion of morality, if you think about it. One could argue that the value prostitutes provide is poison - but then so is what telemarketers sell, too, and prostitutes don't cold-call you to trick you into buying.)
It's not an exclusive relationship. It is a highly transactional ephemeral one.
The basic reproductive qualities of females favor LTR since it takes 9 months at the very least to pop out a kid, during which time mating with other males provides no additional children for them.
It will be an interesting world if these short transactional relationships are basically available anywhere like chips out of a vending machine. I suspect most of the reason for laws against prostitution are a mixture of protectionism for prostitution rings (they need the illegality / high risk to lock in heavy profits) and disproportionately appeals towards the female reproductive strategy that is potentially at a disadvantage with guiltless "normal" sanctioned prostitution on the table at every computer and street corner.
It's a logical evolution of how emotional labor is used in the business world. We pay service and hospitality workers not only to literally do what their job description says but also to be nice to people, keep them engaged and create parasocial (or at least one-sided) relationships that make customers come back and spend more money. Live streamers have long deployed similar strategies with the only difference being they profit off of it themselves rather than being paid through an employer (although many employee moderators, editors and even people to manage their social media accounts so they at best only have to engage with a filtered subset of messages directly if at all).
"The whole thing" in this case isn't just AI. It's "brands".
> That's the whole YouTube content creator business, and it's what it always has been.
I mean, no? That’s the most cynical possible read.
A lot of successful YouTube channels don’t even interact significantly in the comments. For example, in Doug DeMuro’s latest video, I don’t see any comments from his channel in the comments.
You’re also claiming that the YouTube content business is all about exploitation, tricking people into selling them junk. But that’s also a cynical read: lots of creators don’t go down that route.
Using Doug DeMuro as another example, the only recent advertisements he has are for his own car auction site and for Turo, which are both relevant and useful services to automotive enthusiasts.
Some other YouTubers just don’t advertise at all and mostly rely on Patreon, like Technology Connections.
Screens with ads intermixed is only optimized towards attention farming long term.
The renewed interest in long firm on YouTube might be something though.
A post digital addiction Internet is possible - I wonder how many early users of the internet now silence their notifications, maybe even run their screens in greyscale.
I'm an "early" user of the Internet (early to mid 90s, I missed out on BBS's and Usenix but had an internet-connected PC in my home before most people I went to school with did).
And yeah I turn off notifications always on everything. Even my smart phone is usually in Do Not Disturb.
I don't know how much of my online experience shaped this, though. I've got an asperger personality (worded as such because I've not been diagnosed) and unexpected interruptions of any kind drive me insane.
Then there are notifications like they have on LinkedIn and now Facebook where they're more like ads than they are genuine notifications. Things such as "so and so shared a post you might be interested in." These are not designed to notify of you something you wanted to be alerted to, they exist to "drive engagement." Because I don't even like notifications when it's something I care about, these really trigger me to the point where I don't even want to use the "platform" anymore. It's so bad on LinkedIn that I just stopped reading the notifications all together since if you try and turn off all notifcations on LinkedIn .. good luck. You'll spend hours navigating through complex multi-page forms clicking on toggles and then they'll just add some new notification and auto-enable it for you. There's no global "turn of all notifications for all devices" option that I've been able to find.
I don't comment on YouTube videos as much as I used to, because people started comparing the comments sections they are shown for a given video to what their friends and spouses are shown and realized that the comments section is also now algorithmically curated for you based on your viewing and commenting habits. This leads into some serious "dead Internet theory" territory, especially related to this article about OF creators using AI to reply to DMs. Because if two people can see different subsets of comments for the same video, and if YouTube has AI reply features for creators ... how can I trust that the comments I'm reading and replying to are actual humans? Even if they are humans, I don't want to interact with people that just agree with me.
This is a bunch of hand-wringing. If the AI reply is what I would have said anyways; awesome! Finding thumbnails is hard, but if AI comes up with a good one? Fantastic!
None of your interactions except the ones with your friends and family are authentic, and they never were. It's a bit silly to get annoyed over OnlyFans models or some mega YouTube celebrity using AI because you're losing out on "authentic human interaction" - you only mattered to them insofar as you provided them with money to begin with.
I think you miss the part where they fake it being real.
Just like you have a formal paper with CEO signature printed out - you know that guy is not going to sign million of copies.
But I feel offended that they think badly printed signature with pixelation will fool me or will make it somewhat better.
I don’t miss interaction with CEO but I know someone put in effort to fool me.
From all Christmas bonuses and gifts over the years from companies I remember only one where manager of 100 or so people in business unit who actually wrote 100 cards with name for each of us.
Politicians should not only be required to sign the laws they back by hand but to fully recite them without error - anything less means the law isn't important enough to make the books.
> Just like you have a formal paper with CEO signature printed out - you know that guy is not going to sign million of copies.
No, but I assume it would still be considered a valid signature in case of some legal dispute. The CEO may not have signed the document by hand (nor even read it), but the company placing the likeness of CEO's signature in the document signals that the CEO accepts responsibility for it. The CEO is still "in the loop" anyway, they had to personally approve the use of their signature like this.
Which is to say, I consider such "fake signatures" perfectly OK. I just don't consider them as a sign of care or personal interest.
Now, marketing communication that does it, is another story. It's bullshit all the way through, signature included.
> I remember only one where manager of 100 or so people in business unit who actually wrote 100 cards with name for each of us.
Which reminds me - even actually hand-written letters can be fake. Have you ever found a hand-written letter inviting you to a Bible study?
I grew up in a religion that's big on preaching; mostly door-to-door, but when that's for some reason impossible (e.g. time, health constraints), people would write letters instead. Some people were real "high performers" here, in the sense they would sit down over couple evenings and hand-write couple dozen letters, to be distributed around some neighborhood instead of going through it personally. I used to be impressed by dedication, but it eventually dawned on me - it's just exploiting the faux personal connection. They're selling something (which they may feel is genuinely worth it), and hand-written letters is just a sales tactic. They're hoping you pick it up and think about how much effort someone put into a personal letter to you. But the effort is not genuine; it's a fake signal. In reality, the author probably had a good time spending an evening with friends, writing a letter after letter after letter.
So while I 100% believe the intentions of that manager of yours were pure and his heart was in the right place, I post this as a warning for the general case: high effort doesn't automatically imply it's genuine and honest. If it feels like sales, it probably is.
Related: the secret to pulling off a magic trick is to put much, much more effort into preparing the trick than a reasonable person would expect. Same applies to sales.
> Just like you have a formal paper with CEO signature printed out - you know that guy is not going to sign million of copies.
Then don't. It is pathetic and disingenuous to pretend to be personal when you are not. Especially those who spend extra on a single squiggle of blue printer ink.
The point is that people pay to have a genuine interaction. If everything is fake, why interact at all? Imagine you were typing into the void and no one saw what you said. Would you continue doing it if you knew? How about on this forum?
I actually been in this exact scenario before. I and another friend were avidly into Hearthstone, and another third person was playing with an Hearthstone Cheat bot in an University study room. We asked to watch for a while.
After a while, some dread was setting in. We started asking questions:
* Why did it hover that card?
"To pretend it's human. The card has less than 10% playrate on that class"
* Did it... just spam the Well Met meme while going face?
"Of course. Because people do it"
* Wait it ropes the opponents?
"Yes. You can set it to rope back"
We kept seeing more and more behaviors. It would squelch noisy opponents. It would even tap the ground pretending to be a bored person. And then it hit me: 95% of Hearthstone players are bots. Every single human behavior that you could perhaps use to identify 'people', it faked.
It's hard for me to understand how Tinder is not dead yet. One big pile of pop-up ads (even when you pay you get upsell popups) and some Chinese scammer bots.
I'd argue that what Match Group did with OKC was bordering on criminal. Everything on it worked and it worked for small, often marginalized groups. It was turned into a worse version of tender.
It's highly dependent on location. I use it while traveling, and yes in a few countries it's useless, but I've met 300-400 people over the last 7 years. It's added more value to my life than any other single app (even though I've never paid a dime for it).
For the record, I'm male, mid 30's, and average looking.
An average of 1 person per week over a period of 7 years? Impressive. Is it fair to assume you're using a relatively shallow definition of "value" here, or was there something else you had in mind?
Or people might be less lonely. There will come a point where the online experience becomes worthless and people will place greater importance on face-to-face interaction. That's how it was 20 years ago.
Why would you quit hearthstone over the presence of bots?
It's not like you have meaningful interactions with the other players anyway (beyond the occasional post-game friend request which has a 50/50 chance to be abusive).
If you are just playing against bots, the challenge is arbitrary. What would top 10% actually mean? 5000 ELO? Albeit, the ranks are slightly arbitrary already as the matchmaking algorithm significantly influences your competitive experience. But you know when you get higher rank, you have proven you are better than increasing amounts of real players. If every person was actually playing a single player variant of the game (matched against bots), the reward for climbing the ranked ladder is significantly diminished.
Well unlike with modern multiplayer games where the matchmaking algorithm decides who you play with, on the Internet you can still choose what websites to visit.
Honestly, if more people wondered about who'll read what they write online, and whether it matters that those folks read it, the Internet would probably be a better place.
I really don't care about youtubers using tools like this. If it works for them and saves them time: great! If it reduces their level of authenticity: that'll lead to a correction of their popularity.
For OnlyFans... meuh. To me, the idea of interaction there reminds me strongly of the "adult" phone lines of old. You want someone to tease you and say naughty things to you? Well, you can get that for 99 cents/minute.
If you want to buy the attention of someone specific, when they can let someone/something else do that for cents on the dollar... don't be surprised to be tricked.
On only fans people are paying extra for individual specific attention.
But AI can be good enough to fool people that are specifically looking for that experience.
Sure, free market view, if the automated chat isn't good enough then it will lead to people leaving. The point is, it is good enough.
Don't think anybody is complaining that people on only fans are being duped, maybe more concerning that if you can have realistic video and chat, and people are being fooled, then there is some wider impacts on society. Large chunks of people could get fooled by any number of relationships that aren't real.
Hallmark cards tend to leave a lot of blank space to write. It’s absolutely genuine when you grab a pen and start writing. Bonus points if the Hallmark message is relevant to the person receiving it and maybe something you can riff on in the handwritten part. Funny cards are also great!
So I would say the golden rule continues to apply: the more effort is visible, the more genuine it is.
To me, yes. Key point being, "if I review and approve of the message" - if you actually do this, then yes.
Being genuine is all about how much actual care and heart you pour into thing. That's really something only you can truly know. Using generative AI doesn't automatically make it not genuine, much like using a grammar checker or a thesaurus doesn't.
Conversely, not using AI tools doesn't suddenly make YouTube and OnlyFan creators' comments genuine. They never were. There is no care and heart in there, only salesmanship, and it's most likely outsourced to a brand management company anyway.
I recently went to an social interest group meeting. It begins with a presentation, which is generally confirmed in advance by the organizer based on a text abstract of the planned presentation submitted by a future presenter. This time, the presentation turned out to be staggeringly bad, mostly consisting of filler words and interjections, and presenter was struggling to express even the basic idea. I could not fathom how that could happen, considering the idea was presumably expressed in the abstract. I had to write one before, and felt annoying and difficult to write down what I felt was all ready to be said in my head, but as a result I knew what I wanted to say and how I could say it when in front of a dozen of people.
At some point, the organizer (to help situation) quoted from the abstract that was submitted, and it was indeed apparently decently written. After that, I could not help thinking that the likelihood the abstract was written by an LLM is very high. In that case, the presenter certainly reviewed it, but crucially did not write it—the thought process that makes the idea part of your active vocabulary, and you capable of expressing it, would not have taken place.
To reiterate, I don’t know whether it happened or not, but even if no LLM was involved in this instance (perhaps it was just a particularly violent case of stage fright, despite the event being very small and the IRL vibe extremely casual) it would be beyond silly to assume that it would not be happening going forward.
I used to think that it is beneficial if an LLM can help in handling certain boring signaling communication for people who are very bad at it, acting as a sort of an equalizer. A model writes stuff, you approve it, and you gain access somewhere without having, say, the written language flair of someone who went to a prestigious school, or having to spend time on something that seems unnecessary.
I am changing my mind now. Sure, the case I have described is one of the more extreme ones, but it made me think how signals are actually signals for a reason[0], and when some signals go away the communication field does not become equalized—instead, other signals and barriers are used: money, IRL meetings, invitation, some sort of privacy-violating invasive check of humanity, etc., or the communication that relied on some signals before would simply not happen now. When the Web and tech in general had removed a whole lot of constraints on communication, we still could rely on those signals, but that is apparently coming to an end.
Writing a birthday card is another endangered signal. The impression from the movies, how these gestures become very cheap if they come from some rich CEO who certainly has a personal assistant for this sort of stuff, now applies to everyone (including people who would never touch an LLM with a ten foot pole). Once we all know that a birthday card can be reduced to “I have read and approved this message”, as social beings we won’t stop needing the psychological impact of such positive gestures—we only stop receiving it.
I am not sure I see all of the above as positive (even if in the latter case I am slightly optimistic that some viable substitute for those signals could be found within personal relationships).
[0] Even that reason is dubious, like discrimination by a social criteria, well then that’s not going to go anywhere. Tech is not going to solve that human problem, besides perhaps a very fleeting handful of months when some techies gain edge while everyone is still catching on. New barriers will be erected, the core problem left unaddressed.
Is it? Is it also wrong that they so far weren't labeled as [copy-pasted], and/or [outsourced to influencer management agency], and/or [not by ${influencer name}]? YouTube influencers and vloggers are brands, not people; they go big enough, they start outsourcing this stuff, which to my mind is just like "AI slop", except produced by protein parrot instead of silicon one.
Nah, first and foremost, the comment page and the video itself should start with Surgeon General's warning: "You're watching a long-form, semi-interactive ad. None of this is authentic, and none of it is meant to do anything good or nice for you."
(And perhaps also: "You're probably better of going for a smoke instead of consuming this.")
A lot of it is labeled as "you are talking to me personally".
It is akin to a movie stating "a true story" - some liberties may be taken but if the protagonist becomes world president then travels to mars and becomes king of the martians. I am going to start looking for citations.
> Is it also wrong that they so far weren't labeled as [copy-pasted], and/or [outsourced to influencer management agency], and/or [not by ${influencer name}]?
IMO, yes. Copy-pased might be accepteable if it is the author himself doing the selection of what to copy and paste.
If the main value of the comments is interacting with the person in question then anything less is fraud. If authenticity doesn't matter then the label won't impact the desired effect.
Should people know that these interactions have never been genuine? Yes. Is that an excuse for scamming people? Absolutely not.
Expecting internet celebrities to have "authentic" interactions with you is just parasocial relationship. It always has been, and gen-AI just reveals it.
Yeah, and it doesn't even change it - it just makes it more appealing for news to run with a story. Of course, the focus is on AI/authenticity angle; I'd thought they'd give some space for the plight of brand agency marketers (and cheap labor they subcontract to) being pushed out of their jobs by LLMs, but I guess that would require first explaining to people that it was those marketers who people were having their "relationships" with all along. But that's just too sad and complex to report on; it's more of an op ed stuff anyway.
It reduces the cost of the scam siginificantly which makes it available to many more scammers. It's the same with other spam content on the internet - it was possible before but AI makes the problem so much worse.
Or in other wors, yes we should be concerned about the crime syndicate moving into town even if petty theft existed before.
Well for one only someone with severe mental issues would ever consider paying for porn considering how much of it is available. The corollary to that is that anyone making money from the porn industry is taking advantage of people incapable of making rational purchasing decisions.
And the adult industry knows this, which is why sites like Onlyfans empahsize exclusivity and direct interactions. Faking those makes the whole thing even more of a scam.
I think it should be banned too. And maybe they do think that, but I'm just saying, anyone whose livelihood is based on the appreciation of other people should remember not to alienate those people too much
I mean, I guess if your sole goal with consuming content online is to kill free time and pound dopamine out of your brain without putting in any effort, then yeah, fair enough. I'm sure Mr. Beast appreciates your viewership.
I don't follow any creators for such a purpose. I follow creators who make interesting and meaningful things, be they YouTube videos or otherwise. The sorts of people who make a thumbnail that explains what the video is, not just one that's most likely to get attention in the algorithm. Conversely, those folks often respond to people who comment on their interesting and thoughtful things in interesting and thoughtful ways, with interesting a thoughtful replies. This is called a conversation, and it exists for purposes far beyond engagement, and is not a task well suited to AI automation (thank fucking god).
One would counter that if your (if you are indeed some sort of creator) replies are so easily automated, and your thumbnails automated, then the question must be asked... who needs you? How long until YouTube replaces you with a bot, trained on your previous videos, and tells you to kick rocks? After all, you're an unnecessary expense.
> If the AI reply is what I would have said anyways; awesome! Finding thumbnails is hard, but if AI comes up with a good one?
If the AI can replace you so easily then your interactions are worthless. In practice it won't say what you would have said but some corporate approved PC-filtered version of it devoid of any soul.
> None of your interactions except the ones with your friends and family are authentic, and they never were.
Speak for yourself. Not everyone is a sociopath.
> It's a bit silly to get annoyed over OnlyFans models or some mega YouTube celebrity using AI because you're losing out on "authentic human interaction" - you only mattered to them insofar as you provided them with money to begin with.
If (when) this catches on, it won't just be "mega celebrities" using these dystopian methods. But no, even for them expanding the scope of their ability to delude and take advantage of simps is a negative.
There is none, because the whole thing was never good faith. It's always been about tricking people into having a real interaction, to make it easier to exploit them for ad clicks or to sell them junk. That's the whole YouTube content creator business, and it's what it always has been.