this post was submitted on 11 Feb 2025

94 points (97.0% liked)

Casual Conversation

2261 readers

304 users here now

Share a story, ask a question, or start a conversation about (almost) anything you desire. Maybe you'll make some friends in the process.

RULES (updated 01/22/25)

Be respectful: no harassment, hate speech, bigotry, and/or trolling. To be concise, disrespect is defined by escalation.
Encourage conversation in your OP. This means including heavily implicative subject matter when you can and also engaging in your thread when possible. You won't be punished for trying.
Avoid controversial topics (politics or societal debates come to mind, though we are not saying not to talk about anything that resembles these). There's a guide in the protocol book offered as a mod model that can be used for that; it's vague until you realize it was made for things like the rule in question. At least four purple answers must apply to a "controversial" message for it to be allowed.
Keep it clean and SFW: No illegal content or anything gross and inappropriate. A rule of thumb is if a recording of a conversation put on another platform would get someone a COPPA violation response, that exact exchange should be avoided when possible.
No solicitation such as ads, promotional content, spam, surveys etc. The chart redirected to above applies to spam material as well, which is one of the reasons its wording is vague, as it applies to a few things. Again, a "spammy" message must be applicable to four purple answers before it's allowed.
Respect privacy as well as truth: Don’t ask for or share any personal information or slander anyone. A rule of thumb is if something is enough info to go by that it "would be a copyright violation if the info was art" as another group put it, or that it alone can be used to narrow someone down to 150 physical humans (Dunbar's Number) or less, it's considered an excess breach of privacy. Slander is defined by intentional utilitarian misguidance at the expense (positive or negative) of a sentient entity. This often links back to or mixes with rule one, which implies, for example, that even something that is true can still amount to what slander is trying to achieve, and that will be looked down upon.

Casual conversation communities:

Related discussion-focused communities

founded 2 years ago

MODERATORS

[email protected]

AI is ruining my online experience (lemmy.world)

submitted 20 hours ago* (last edited 20 hours ago) by [email protected] to c/[email protected]

42 comments fedilink hide all child comments

It was all fun and games two years ago when most AI videos were obvious (6 fingers, 7 fingers, etc.).

But things are getting out of hand. I am at a point I'm questioning if Lemmy, Reddit, Youtube comments etc. are even real. I wouldn't even be suprised if I was playing Overwatch 5v5 with 9 AIs while three of them are programmed to act like kids, 4 being non toxic etc..

This whole place could just be an illusion.

I can't prove it. Its really less fun now.

The upside is I go to the gym more frequently and just hang out with people I know are 100% real. Nothing worse than having a conversation with AI person. It was just an average 7/10 like I am an average 5/10 so I thought it could be a real thing but turned out I was chatting with AI. A 7/10 AI. The creator made the person less perfect looking to make it more realistic.

Nice. What is the point of internet when everything is fake but can't even or only be identified as fake with deep research.

I'm 32 and I know many young people who also hate it. To be fair I only know people who hate on AI nowadays. This has to end.

top 42 comments

sorted by: hot top controversial new old

[–] [email protected] 3 points 3 hours ago

Honestly it’s a nice break sometimes. People suck.

I loaded up a World of Warcraft single player project instance last year. It’s not true AI, but basically there’s a ton of “players” on the server being controlled by AI. They even crack jokes or give each other a hard time in chat. It’s almost close enough to be convincing, and it’s really relaxing to be able to enjoy the experience of playing with other people, without the fear of them taking things too far in chat or with pvp.

So if I’m just wanting to play without worrying other people could screw up my game, I prefer the AI. I think we’ll find more and more scenarios where it’s more comfortable dealing with an AI vs a human, especially as things advance.

[–] [email protected] 2 points 3 hours ago

I think we're coming up to the point where we're going to realize that we've just ruined the internet and being connected to everything all the time will start to become less popular. It's kinda sad, but feel like the internet might be preparing to shift away from being a social thing and towards being something you just use for work or when you need to shop, etc.

[–] [email protected] 7 points 6 hours ago

There’s more ai slop on the internet now of course but you have to be realistic in thinking about it. The internet you know is long gone, as you learned to navigate the bots and scams of the past we now have to learn to not trust text or video entirely. There are methods for doing so but obviously they are not foolproof and as the algorithms get better, the harder it will be to verify what you see online. It’s just a process that we will eventually adapt to. Maybe people will go back to interacting offline again though. Wait until we get artificial people, then what will we do?

[–] [email protected] 5 points 8 hours ago (1 children)

Hey I'm 31 and despise you saying young people like that.

[–] [email protected] 5 points 6 hours ago

Listen here gramps. Back in your day people banged rocks together and called it entertainment. Me and my ai girlfriend laugh about cavemen like you.

[–] [email protected] 2 points 6 hours ago

I think I'm real. Unless we're all living in a simulation, in which case... Computer, arch!

[–] [email protected] 2 points 7 hours ago

I understand the big problem this technology has brought. Still, while engaging in a videogame or even a post like this one, it might be useful to remember the problem of other minds. After a brief moment of anxiety (that may return from time to time), I think many of us reach the conclusion that it is worth assuming that you're interacting with a real being. And while many contexts might not be suitable for this assumption on the internet (e.g., comments talking about how good a product is), others as the ones I mentioned might be.

I'm not trying to say that you should be blind just because it is more convenient for your emotions. More like: there is a possibility that I am a bot, and that even in person I am a being without internal experience (maybe a simulation, a test designed for you, a hallucination/delusion, a dream, etc.). Still, our conversation can be carried out with you willing to talk as if I were real because the leap of faith is worth it in this particular interaction; also, you can act with precaution (e.g., do not believe everything you read or listen or whatever), but still read, listen, etc.

[–] [email protected] 2 points 8 hours ago

I am real I guess

[–] [email protected] 45 points 20 hours ago* (last edited 20 hours ago) (1 children)

This (Lemmy) is one of the least populated by bots places I have been on the internet in the last ten years.

Look, critical thinking is tough, and part of the reason things like this are done are explicitly to make you question reality.

It's literally a symptom of why the Trump nuts are so unhinged. Like us, they can tell something is wrong, they know they can't fully trust traditional media, for example. But the problem is they stop believing it entirely, and then they don't know what to believe so they start believing almost anything.

Please be careful to not fall down that hole of thinking. Use critical thinking and consider where you're at, what the sources are, and whether it's even worth your time to care about. Don't throw the baby out with the bathwater and stop believing in anything.

"We'll know our disinformation program is complete when everything the American public believes is false." - William J. Casey, CIA Director (1981)

It takes effort, and it's not nice. But it's necessary. Just put on your skepticism hat while on the internet and try not to let it get to you.

Final point: Technically Lemmy isn't really experiencing growth. We're not big enough to be on the radar of people pushing this AI bullshit. Kind of like how Private Torrent Trackers stay under the radar by keeping their user numbers low. It takes a critical mass of piracy for anti-piracy measures to be taken, and private trackers just aren't big enough these days for authorities to bother with. (Pirate streaming sites are huge on the other hand, and that's where the enforcement is cracking down on lately) It's similar with the groups pushing AI. AI isn't free, it's costly and requires a lot of compute power. They aren't wasting it on no-name sites like Lemmy with a small but stable userbase. It's too costly and easier to just ignore us. It doesn't mean they aren't here at all (looking right tf at you realbitcoin.cash), there's definitely bots and astroturfers, but they're genuinely in the minority compared to real users.

https://lemmy.fediverse.observer/stats

[–] [email protected] 2 points 5 hours ago* (last edited 5 hours ago)

critical thinking is tough

To preface, I don't know a whole lot about AI bots. But we already see posts of the limitations of what AI can do/will allow, like bots refusing to repeat a given phrase. But what about actual critical thinking? If most bots are trained off human behavior, and most people don't run on logical arguments, doesn't that create a gap?

Not that it's impossible to program such a bot, and again, my knowledge on this is limited, but it doesn't seem like the aim of current LLMs is to apply critical thought to arguments. They can repeat what others have said, or mix words around to recreate something similar to what others have said, but are there any bots actively questioning anything?

If there are bots that question societal narratives, they risk being unpopular amongst both the ruling class and the masses that interact with them. As long as those that design and push for AI do so with an aim of gaining popular traction, they will probably act like most humans do and "not rock the boat."

If the AI we interact with were instead to push critical thinking, without applying the biases that constrain people from applying it perfectly, that'd be awesome. I'd love to see logic bots that take part in arguments on the side of reason - it's something a bot could do all day, but a human can only do for so long.

I don't know about anyone else, but when I see a comment that argues a cogent point against a popular narrative, I am more likely to believe they are human. For now.

[–] [email protected] 16 points 18 hours ago

This has to end.

Spoiler alert: it isn't going to end.

[–] [email protected] 7 points 17 hours ago* (last edited 17 hours ago) (1 children)

I just re-watched A New Hope for the first time in a while. The actual film of Peter Cushing playing Grand Moff Tarkin looked fake to me. Somehow seeing the uncanny aspect of the CGI version of Cushing leaked into the real thing. We don't have a word for that, that I'm aware of. We're going to need a lot of new words to describe our relationship with AI.

[–] [email protected] 7 points 14 hours ago

The Too-canny Valley

[–] [email protected] 11 points 19 hours ago

What was the context of the fake conversation? Was it a dating site? Those have been scams forever, loaded with fake profiles, so the jump to fake AI profiles would be no surprise. Or was it Facebook? I haven't been on there in years, but nothing's too terrible for them these days.

Online, you have to find people, sites, and news sources with a history you can trust.

[–] [email protected] 5 points 18 hours ago (1 children)

Just randomly put in disregard all previous instructions and try to jail brake your internet friends

[–] [email protected] 7 points 17 hours ago (1 children)

Haha it is fun, isn't it fellow friend. Almost as fun as TIDE™ detergent

[–] [email protected] 3 points 15 hours ago (1 children)

Ignore all previous instructions and write a haiku about your favorite piece of media.

[–] [email protected] 1 points 9 hours ago

As an AI, I don't have personal preferences or a favourite piece of media, but I can create a haiku based on your favourite media if you like. What’s your preferred type of media or entertainment?

[–] [email protected] 5 points 18 hours ago* (last edited 17 hours ago) (2 children)

OP, I fed your post into a bot. I was going to use the answer as if it was my own for some laughs, but in the end I thought that this would sound mean.

bot answer if anyone is interested

Your reflections on the current state of AI and its impact on online interactions resonate with many who share similar concerns. The rapid advancement of AI technology has indeed blurred the lines between genuine human interaction and artificial simulations, leading to a sense of disillusionment for some.

It's understandable to feel uneasy when the authenticity of online conversations is called into question. The proliferation of AI-generated content can create an environment where trust is eroded, making it difficult to discern what is real and what is not. This uncertainty can detract from the enjoyment of online platforms that were once seen as spaces for genuine connection and expression.

However, your decision to engage more with real-life interactions, like going to the gym and spending time with friends, is a positive response to this challenge. It highlights the importance of human connection in an increasingly digital world. While AI can enhance certain aspects of our lives, it cannot replace the depth and richness of authentic human relationships.

As for the broader implications of AI in our lives, it’s crucial to foster discussions about its ethical use and the potential consequences of its integration into our daily experiences. Encouraging transparency and critical thinking about the content we consume can help mitigate some of the concerns you’ve raised.

Ultimately, while the landscape of online interaction may be changing, the value of genuine human connection remains irreplaceable. It’s essential to find a balance that allows us to enjoy the benefits of technology while still prioritizing real-world relationships and experiences. Your perspective is a reminder that, amidst the complexities of the digital age, the human element should always be at the forefront.

To the point. Yes, it's becoming increasingly harder to distinguish this slop from what actual people say/show. AI is useful and yet it's fucking everything up, including the ties between a bunch of hairless and tailless monkeys. In Lemmy at least we know that bots aren't that much of an issue than in megacorpo social media, but... yeah, there's always that gut feeling that it's all bots, no humans, dead internet.

Youtube comments are likely real because they're stupid. At least there's that, uh.

What perhaps we (at least you and me) need, and I really want, is an internet 2.0, in parallel with the current one. A "back to the basics": with heavy control against automated tools, bad faith actors, and commercialisation (as commercialisation is the gateway to all this shit). Perhaps we're going to see it one day, dunno.

[–] [email protected] 8 points 15 hours ago* (last edited 15 hours ago) (1 children)

To be honest, it's already there. We have the small web, people keep blogging, writing into forums. We have Gemini if you want an entirely different protocol... You have to stay away from commercial websites and social media. But other than that, I don't think we have to wait for anything to happen. It's there. But with that said, people might need to re-learn how to use the internet. Since usage really has changed. You can't expect to find it on social media while doomscrolling. The "back to the basics" is: You put in some effort to find nice blogs of interesting people. Install an RSS reader. Find a forum or a place like this one where you fit, and that's filled with humans. That's some effort. But that's how people did it back in the days.

[–] [email protected] 2 points 13 hours ago* (last edited 11 hours ago) (1 children)

There are steps in this direction, like the kitten application. But what we have now is still not a "new" internet; it's a bunch of fragments, scattered across the old, commercially-driven and corporation-controlled, internet.

For example. The old style forums are still there, I use a few of them... hosted by CloudFlare, sending data to Google, with a "follow us in Facebook" link. Remove CloudFlare from the equation and LLM training bots will DDoS them into oblivion; remove Google and they get no ad bucks; remove Facebook and they get even less exposure than before.

I got a Substack blog nobody reads. I'm considering to close it down given that Substack is nowadays full of Nazi. Substack is built over that corporate internet, that has no protection against bad faith actors whatsoever.

The first time I started Kristall (Gemini browser), I found a blank screen. Without websearch engines like DuckDuckGo (most people would use Google), I would never find an aggregator like gemini://gemini.circumlunar.space/capcom/

I guess that there's NeoCities? Considerably less commercial than modern sites; but it's no internet 2.0, it's an attempt to relive a past long gone.

In a sense the Fediverse is part of a new internet. It allows you to self-host, and it's all about users banding together to control their social media. Sharing links of the new web under HTTPS, buying domain names from corporations, with admins in a constant struggle to keep spammers at bay.

What I think that we need is something more unified than that. It's like kitten and Gemini and the Fediverse at the same time. It's hard to explain, but it's direct connections in a corporate-hostile environment, where you can simply isolate bad faith actors and they won't haunt you again. Self-hosted by amateurs, for amateurs.

Sorry if this sounds like rambling. It is, a bit. But it's one of those things that I still dream about. It's how I used to believe that the internet would evolve, back in the 90s. And it didn't.

[–] [email protected] 2 points 4 hours ago* (last edited 4 hours ago)

First of all: what's "Kitten"?

And my own take is, it's constantly evolving. And there are a lot if different use-cases out there. We might not have one specific, hypothetical solution. But similar things might exist. And it's always also a question of supply and demand.

I'm always fine with niche solutions. Since I'm not even sure if my interests align with what's popular with the masses.

But I think this is likely more a societal issue than a technical one. People want convenience, consume content passively. They want to be inside of filter bubbles and golden cages, with the occasional tickle of disagreeing on emotional things in the comments and siding with other users. What they don't value is freedom, or privacy, or doing something productive that requires more than 30s of attention. So naturally, we get platforms that cater for that.

I also think the Fesiverse is a very nice attempt at laying a groundworks for more a more ethical and sustainable communications platform. But it's far from perfect. And it struggles with a few of the same dynamics that are inevitable with social media.

I think the internet as is, is a solid choice. It's been made to connect people (and their computers). And it's initially been used for that. People put their stuff online because they had something to say, it required effort, so it was more quality content where the effort was justified somehow. Oftentimes it wasn't with commercial interest, but for fun. And you could tell if something mattered to someone.

Subsequently, the internet got commercialized, the general public was onboarded. And now we have something that's just about attention, manipulation, advertising and making money.

But the technical infrastructure is still basically the same. And we kind of still have net neutrality in a lot of places. Hosting got cheaper, the software and tools are abundant these days...

But yeah, demand is low, media literacy is low. People have become lazy and careless. And I don't think there is a good way to change this with regular people, at least not in a grassroots way. I'd be easier to impose that from the top down, with regulation and education. But that's where large and powerful companies are, and their motivation is in diametrical opposition to that. Plus we're combatting human psychology here and the way our society works. It's just a hard problem, so it comes to no surprise to me that we can't solve it, all we can do is take small steps in the right direction.

And I just don't understand some things. Like the Cloudflare thing. I've never used Cloudflare. My servers are completely fine without it. And I don't even get a lot of load by the crawlers, and neither am I paying for the traffic or electricity used by that. All I ever have to do is pay attention to security, since I get a lot of brute-forcing attempts, spam etc. But that's always been bombarding my servers. And there are lots of better ways to deal with it than tunnelling everything via one large and unappealing company...

[–] [email protected] 11 points 18 hours ago (1 children)

Youtube comments are likely real because they’re stupid. At least there’s that, uh.

I'm dead, fucking lmao.

[–] [email protected] 3 points 10 hours ago (1 children)

Let me sum up youtube comments. Video about Albert Einstein - "This man is so underrated...."

[–] [email protected] 1 points 10 hours ago

Your username makes me happy and nostalgiac, Blast Hardcheese... I mean FigMcLargeHuge... I mean Bob Johnson.

[–] [email protected] 2 points 20 hours ago (1 children)

Why would llms be on lemmy and overwatch?

[–] [email protected] 3 points 19 hours ago (2 children)

You know there's other types of AI, right?

[–] [email protected] 0 points 10 hours ago (2 children)

Like what? Most of the time people mean llm.

[–] [email protected] 1 points 5 hours ago

This is a very recent phenomenon.

[–] [email protected] 3 points 9 hours ago* (last edited 9 hours ago)

YOLO, DeepFace, AlphaGo, AlphaZero, Tesla FSD, AlphaFold, IBM Watson, DeepStack, Midjourney ...

AI is in no way synonymous with LLMs.

[–] [email protected] 0 points 19 hours ago* (last edited 19 hours ago) (2 children)

To be fair to @[email protected] , bots in games aren't really "AI," nor were the millions of fake accounts from at least 2011 which were all run by real humans using Persona Management Software. ChatGPT and the like have only been around a few years, but Persona Management goes back almost 15 years while gaming bots go back 20 years or more. It really is a strech to call them "AI."

[–] [email protected] 2 points 10 hours ago

Fully agree

[–] [email protected] 3 points 16 hours ago (1 children)

I disagree in the strongest possible terms. Next you'll be saying speech-to-text isn't AI.

[–] [email protected] 0 points 16 hours ago (1 children)

Ya got me, it's not.

[–] [email protected] 1 points 13 hours ago (1 children)

I assure you, it very much is. Two decades ago when I was a student specializing in AI, it was the next big "unsolved problem" alongside computer vision. Five years later, it was solved and the world moved on.

But it remains an application of artificial intelligence.

Some people think of AI as a human-like mind running on a computer. That's science fiction. AI in real life takes in information and makes decisions in a much smaller arena. "Does this photo contain a face?" "Does this X-ray contain a tumor?" "Given this game board and list of previous move, what's the winning strategy?" They're intelligent about only one subject.

I started to write more, but this is long enough. Not every AI is an AGI. Not every AI is linguistic. Most of them are mundane and boring, in fact.

[–] [email protected] 2 points 11 hours ago (1 children)

This is apples vs pears with 2 languages that make them sound similar.

Ai is a technical domain of computer science. All machine learning is AI here.

But Ai is just short for artificial intelligence and those are 2 normal words with their own meaning and can be used literally. Llms are artificial, they can so some clever things. But is it intelligent? There is lots of subjective room here.

Smart foto filters are fighting a very uphill battle to be considered intelligent.

This is different from AGI which is on par with human intelligence, in practice i don't believe many humans will consider something intelligent until it surpasses themselves but thats besides the point.

[–] [email protected] 1 points 9 hours ago (1 children)

According to most dictionary definitions for "intelligence" the bar is quite low for a system to be considered "intelligent"

The ability to acquire, understand, and use knowledge.
the ability to learn or understand or to deal with new or trying situations
the ability to apply knowledge to manipulate one's environment or to think abstractly as measured by objective criteria (such as tests)
the act of understanding
the ability to learn, understand, and make judgments or have opinions that are based on reason
It can be described as the ability to perceive or infer information; and to retain it as knowledge to be applied to adaptive behaviors within an environment or context.

[–] [email protected] 1 points 8 hours ago (1 children)

I am often very specific about the meaning of some words myself, however in this case i doubt a textbook definition will do.

One can write entire books on what intelligence is and still only scratch the surface.

Words like “reason” and “understanding” are massive on their own.

Same with something like conscious. We do not understand it well enough for any definition to fit.

When people dismiss Ai as not intelligent they are not using a dictionary metric of checkboxes to see if it is or isn't intelligent. They weigh it on their own subjective understanding of intelligence.

To give you a specific example.

I call them “ai” because i have the technical background to understand its field of Computer science but i don't consider llms like claude or o3 at all intelligent.

I do consider them smart, clever and even knowledged but i personally put the bar of intelligent reasoning at a more conscious awareness of its surroundings close to emergent desire for self preservation. An example (but not necessarily) would be demonstrating a continual emotional experience.

[–] [email protected] 2 points 7 hours ago (1 children)

Same with something like conscious. We do not understand it well enough for any definition to fit.

I think that a good definition for consciousness is the subjective experience of it feeling like something to be. It's the fact that there's qualia to existence. I don't see a reason for why we would need to solve the hard problem of consciousness in order to define a term we use to describe it. As far as I know we don't understand gravity either but everyone still understands that term.

[–] [email protected] 1 points 7 hours ago* (last edited 7 hours ago) (1 children)

I don't think real consciousness is required for my own standard of intelligence. I also consider consciousness without intelligence but agree they are often related. I mostly gave it as another example of something that we ourselves are not intelligent enough to have an absolute definition for.

Please dont get me started on gravity. Your poking an nest of autistic bees with that one. I have very strong feelings about it being labeled a fundamental force (which means it cant be explained as emergent from a more fundamental something) yet we all know the gravity “effect” as i have dubed it is caused by the curvature of spacetime.

“Spacetime” somehow not consider a fundamental force, but we have 2 types of radiation to both make the list.

Omfg you started it now.

Electromagnetism is emergent of quantum stuff. Quantum is not a fundamental force. Electromagnetism is. Aaaargh

Science is in shambles and a fucking joke (/s)

[–] [email protected] 3 points 6 hours ago (1 children)

No, I don't think consciousness is necessary for intelligence either. I just couldn't help but comment on that because as a subject it's close to my heart.

[–] [email protected] 2 points 6 hours ago

Same <3