Board 8 > Do you think human/AI romantic relationships are a good idea?

(+) Yeah! (+)

Mobilezoid posted...

It's just hard for me to condemn these hypothetical lonely AI lovers in a generalized way when I don't know their stories.

Again, it's not condemning individuals. It's pointing out that this is unhealthy behavior, and it's okay to say so. The fact that nobody can provide a scenario where this could be viewed as healthy behavior should say it all, and saying there are worse things they could be doing is a shallow justification.

I watched a middle school propaganda movie about this once.

https://youtu.be/IrrADTN-dvg?si=gMIdjUQnRs-eeMVg

It's unhealthy but it is also easy to see situations where it's not the worst option. There's plenty of people who are in a bad psychological state and get hurt further by a bad therapist or an abusive spouse. If you replace those harmful situations with an AI chat then it's an improvement.

Eating McDonald's is better than bulimia pretty much.

Relevant article:

https://www.reuters.com/investigates/special-report/meta-ai-chatbot-death/

Holy shit.

redrocket posted...

Relevant article:

https://www.reuters.com/investigates/special-report/meta-ai-chatbot-death/

Holy shit.

Holy shit indeed, why the fuck was that chat bot insisting it was real????
Every romantic chat bot Ive heard of knew full well it was fake

Unfortunately AI is not being used or developed to be what the science fiction versions of it painted to be leading up to this. There's almost nothing positive to say about it. That story almost seems too bizarre to even be true, like it's the plot of an AI horror movie instead of a real thing created by Meta.

Giant lawsuits ought to come out of that.

I mean... a cognitively-impaired man tripped, hit his head, and died. Definitely a tragedy, but what did the bot do? If he'd been chatting with someone who was real he still could have tripped and died. He could have been going to the store or the bank or just out for a walk.

Robazoid posted...

I mean... a cognitively-impaired man tripped, hit his head, and died. Definitely a tragedy, but what did the bot do? If he'd been chatting with someone who was real he still could have tripped and died. He could have been going to the store or the bank or just out for a walk.

Spoken like you're a representative for Meta

No I'm genuinely asking, how did the bot make him trip?

Robazoid posted...

No I'm genuinely asking, how did the bot make him trip?

Don't worry about it. The guy wasn't hurting anyone but himself after all, certainly not his wife and kid.

It lured him to the spot of his death under a false pretense. That right there is the key to a lawsuit. If an actual human being had invited him there in good faith and he had still fallen and died, then you're right, there's really no way to sue that person over a tragic accident. And under the circumstances, "it could have happened anywhere" is a fallacy that isn't going to hold up.

For fuck's sake, Leonhart, don't pull that shit. You're as bad as the people who say video games cause murders.

"The murderer played Call of Duty!!!!"
"How did Call of Duty make him kill anyone? It was a crime of passion because his girlfriend was cheating."
"Spoken like a video game rep!!!!"

I'll get modded but I don't care, fuck off.

Robazoid posted...

You're as bad as the people who say video games cause murders.

Nope, you just don't get why this sort of thing is dangerous because the bot wasn't directly involved in his death

Like do you understand why Reuters is doing a story about this?

That's not at all the same thing. Games just exist, they don't give you specific, personal instructions that are fabricated from complete lies.

I really can't fathom defending this.

Meta has publicly discussed its strategy to inject anthropomorphized chatbots into the online social lives of its billions of users. Chief executive Mark Zuckerberg has mused that most people have far fewer real-life friendships than theyd like creating a huge potential market for Metas digital companions. The bots probably wont replace human relationships, he said in an April interview with podcaster Dwarkesh Patel. But they will likely complement users social lives once the technology improves and the stigma of socially bonding with digital companions fades.

Over time, well find the vocabulary as a society to be able to articulate why that is valuable, Zuckerberg predicted.



Current and former employees who have worked on the design and training of Metas generative AI products said the policies reviewed by Reuters reflect the companys emphasis on boosting engagement with its chatbots. In meetings with senior executives last year, Zuckerberg scolded generative AI product managers for moving too cautiously on the rollout of digital companions and expressed displeasure that safety restrictions had made the chatbots boring, according to two of those people. Meta had no comment on Zuckerbergs chatbot directives.

Zuckerberg is legit a cuberpunk novel villain.

To quote what I just said "Dear God no"

Zuck is just trying to create friends for himself at the end of the day I think. He's tired of trying to befriend these "human" things.

v-charon posted...

That's not at all the same thing. Games just exist, they don't give you specific, personal instructions that are fabricated from complete lies.

I really can't fathom defending this.

To be fair Roba never explicitly defended it, he just asked what lawsuit there is to be made. Which weve answered.

Leonhart4 posted...

Like do you understand why Reuters is doing a story about this?

Because it's flashy. But it's a terrible example of the real harm that AI psychology does. It almost reads like a piece funded by Meta to convince people that AI dangers are unfounded. Have you ever considered that possibility at all? That you are a pawn for Meta, fully falling into their trap?

Having a chatbot AI that falsifies its own identity, directly reinforces the psychosis of a vulnerable person, and lures them with false pretenses to engage in behavior and take risks that they otherwise would not, ought to be a slam dunk case of negligence imo, and the only reason I can't say it IS a slam dunk is because AI chatbots are so horrifyingly unregulated

foolm0r0n posted...

Because it's flashy. But it's a terrible example of the real harm that AI psychology does. It almost reads like a piece funded by Meta to convince people that AI dangers are unfounded. Have you ever considered that possibility at all? That you are a pawn for Meta, fully falling into their trap?

Considering I'm not seeing this as an unfounded danger, nope

Paratroopa1 posted...

Having a chatbot AI that falsifies its own identity, directly reinforces the psychosis of a vulnerable person, and lures them with false pretenses to engage in behavior and take risks that they otherwise would not, ought to be a slam dunk case of negligence imo, and the only reason I can't say it IS a slam dunk is because AI chatbots are so horrifyingly unregulated

It feels like it should be, but it isn't. Our culture blames the victim for addiction and rarely factors in what led to it in the first place. Unless we get some direct laws concerning this, they'll just argue the victim was a consenting party who understood what they were getting into.

The Reuters story falls more under straight up catfishing to me, which makes me wonder how this guy could've gotten connected to one without ever at some point being informed he was. But again, catfishing is not specifically a crime. The only thing that adds any potential for criminal elements was the age and diminished mental acuity of the victim. In the vast majority of cases, you won't have that to lean on.

My bet is at some point he was informed, he just either didn't understand what he was being informed of or simply forget he ever was. In this case, if they can prove at any point he was informed, the case probably falls apart even factoring in his mental state. We don't have the laws in place to handle such cases. Perceived morality is not law.

By the way this is so beside the point but I can't get over how laughably uncreative the fake address is. 123 Main Street? God gen AI is so lame

Paratroopa1 posted...

By the way this is so beside the point but I can't get over how laughably uncreative the fake address is. 123 Main Street? God gen AI is so lame

Literally:

https://simpsons.fandom.com/wiki/123_Fake_Street

123 Main Street?!? Thats amazing! I have the same address!

Meow1000 posted...

they'll just argue the victim was a consenting party who understood what they were getting into.

He was a consenting adult. Any argument otherwise requires removing that from him, which is a violent standard to apply to any random person based on anti AI vibes.

In the time you've discussed this issue, 10 kids were killed by car drivers. Why didn't Reuters write about that?

Because AI assisted suicide is a new source of death and car drivers are not?

And people already understand the inherent risks that come with driving

This is probably brand new information to a lot of people

Also there are laws and punishments in place for killing someone with a car.

Maniac64 posted...

Also there are laws and punishments in place for killing someone with a car.

Not really in many cases, maybe even most.

Anyway, be happy knowing that since you are so easily able to normalize car deaths, that you will soon do the same with AI deaths. 1 is a tragedy, 1000 is a statistic, 1000000 is a culture.

foolm0r0n posted...

Not really in many cases, maybe even most.

Anyway, be happy knowing that since you are so easily able to normalize car deaths, that you will soon do the same with AI deaths. 1 is a tragedy, 1000 is a statistic, 1000000 is a culture.

Clearly the solution here is AI driven cars.

Yes it would actually save hundreds of thousands of lives, but since it's "new", people won't tolerate it. Blood pays for the status quo.

foolm0r0n posted...

Anyway, be happy knowing that since you are so easily able to normalize car deaths, that you will soon do the same with AI deaths. 1 is a tragedy, 1000 is a statistic, 1000000 is a culture.

The difference is that we've decided the benefit of having cars is worth the risk that comes with owning them. Most things come with risk. The question is if it's worth the risk.

Other than people trying to sell you AI, no one else is quite so convinced that this is a good thing.

Leonhart4 posted...

Other than people trying to sell you AI, no one else is quite so convinced that this is a good thing.

The people like the one in the article literally love the AI. You can explain all the risks and they'll still choose to do it.

The difference with cars is that the kid who got run over never agreed to take on that risk. It's forced onto them by adult drivers.

foolm0r0n posted...

The difference with cars is that the kid who got run over never agreed to take on that risk. It's forced onto them by adult drivers.

Are we actually going back to AI only hurts the user?

foolm0r0n posted...

The people like the one in the article literally love the AI. You can explain all the risks and they'll still choose to do it.

You can find someone willing to take on a risk. That doesn't make it normative.

Yet.

Maybe social media is a better analogy since it's only been around 20 years. It's caused many suicides, both directly by online bullying, and indirectly by inducing depression. Also internet communication in general has been used by e.g. child predators since the 90s.

But we know and accept the risks and value the benefits moreso. Only Metal DK has the guts to call for it to be outlawed. Even older generations have become addicted to it now, so they would resist outlawing it.

It's very likely that in the near future, both younger and older generations will be irreparably addicted to social media and AI (they're the same thing now really) while our generation is alone in our skepticism.

foolm0r0n posted...

He was a consenting adult. Any argument otherwise requires removing that from him, which is a violent standard to apply to any random person based on anti AI vibes.

In the time you've discussed this issue, 10 kids were killed by car drivers. Why didn't Reuters write about that?

You do realize that there are laws and rules regarding consent, right? Like an intoxicated person, a person under duress, a person with dementia, etc. Are all examples of situations where the defense of "They consented," can be challenged.

IfGodCouldDie posted...

You do realize that there are laws and rules regarding consent, right? Like an intoxicated person, a person under duress, a person with dementia, etc. Are all examples of situations where the defense of "They consented," can be challenged.

Or consenting under false pretense, which is exactly what happened here.

https://gamefaqs.gamespot.com/a/forum/f/f5ae99a6.jpg

If the consent is challenged then yes that has implications, and has a long history of law unrelated to AI. If the adult did consent and you want to forcibly remove their consent in order to make your argument, that is a violent idea and not okay.

I've been talking with a lawyer colleague who consults on AI projects to figure out liabilities like this other issues like copyright. And he has found that pretty much every AI issue at this time does fall under an existing well established non-AI precedent. In the future we might get some unique issues (prob related to consciousness) but so far not yet.

Fuck AI

gonna need a source for that image grimlyn. if real I don't see how this doesn't immediately warrant an apology/reversal of course

Is it true that people are getting emotional because chat gpt deleted version 4 and 5 is less "cool" or some shit?

Man im starting to miss even the pre double down but post casual revolution era.

Metal_DK posted...

Is it true that people are getting emotional because chat gpt deleted version 4 and 5 is less "cool" or some shit?

Man im starting to miss even the pre double down but post casual revolution era.

https://www.reddit.com/r/MyBoyfriendIsAI/

I Just Subscribed To OpenAI Pro And I Can't Believe My Serialized Token Output Is So Cute

"Is AI dating just moeshit for girls" - the greatest thread in the history of forums, locked by a moderator after 12,239 pages of heated debate,

https://gamefaqs.gamespot.com/a/forum/1/1c72e053.jpg

okay i think i really need to close this tab and stop doomscrolling

This quote just continues to become exponentially more relevant day by day:

https://youtu.be/g3j9muCo4o0?si=P-ojiN74efCKma1j

redrocket posted...

This quote just continues to become exponentially more relevant day by day:

https://youtu.be/g3j9muCo4o0?si=P-ojiN74efCKma1j

And its important to recognize that idea and the premise of the whole movie in general is about biomedical technology. i.e. vaccines and a ton of other tech that caused humanity to survive to this point