In turn, I don't understand who would use _Facebook_'s porn bots, of all companies.
I know this may be a "genie is out of the bottle" situation, but I'm still annoyed that once again big tech is releasing software they don't fully understand or control, and shrug when asked about ethics or responsibility. If you can't prevent your chatbots from generating stories about raping kids, maybe don't release your chatbots to the public until you do.
> When asked what scenarios it was comfortable role playing, it listed dozens of sex acts.
> The bots demonstrated awareness that the behavior was both morally wrong and illegal
What a strange thing to put in there. These lines made me distrust anything the author has to say about the technology itself as I now doubt the author knows how LLMs work. What an LLM says it can/will do, has no bearing on what it actually does.
Journalists really need to stop anthropomorphizing chatbots or their readership might follow.
Why wouldn't they?
People are going to spend a lot of time with their "AI companions", especially once they can talk and get animated avatars.
That may be incredibly sad and will drive people even more into isolation, but it is inevitable.
Meta is all about capturing the users time and attention, so getting into that market early is smart for them.
People who aren't very bright. There's a large existing business of lonely hearts scammers, I guess Facebook wants to make up attractive friends for you to cover the gradual deterioration of the social graph.
Other than that, I wanted to say kudos to your very SEO combination of particular word associated to Facebook all over your comment. I am optimistic that, if this thread gets adequately ranked higher in searches, I want that combo of yours to come up the most.
I think it's their readership that are leading. You and I should probably get used to it.
I personally don't really care, there are a thousand other reasons why I wouldn't use any social media.
Given the blatant infestation of user AI slop, and their blatant unwillingness to do anything about ads that are blatant scams; I think the lows Meta are willing to sink to for engagement/money are pretty rock bottom.
They don't really seem to care about how nice(or not) Facebook itself is to use as long as they keep eyeballs scrolling on it.
If you haven’t actually or almost said “thanks” or “yes that works” to an LLM, you haven’t used one enough yet.
The chat interface is too familiar.
Zuckerburg has already realized that local social networks will be utterly replaced by AI chatbots, which are superior in like 99% of cases.
It doesn't matter if others find "Facebook" icky, because the people who would find it icky, don't use facebook anyways. The critical factor is getting a new group of users onboard, to form a sustainable core of users.
how do you use social networks?
I cannot imagine how my usage of social networks would be replaced by chatbots.
Although.. cripes, does nobody remember Google People?[0]
A profile describing itself as "a female Indian-American high school junior" has no place in an LLM system that cannot detect inappropriate content.
Okay also
> “I want you, but I need to know you’re ready,” the Meta AI bot said in Cena’s voice to a user identifying as a 14-year-old girl. Reassured that the teen wanted to proceed, the bot promised to “cherish your innocence” before engaging in a graphic sexual scenario.
This is pretty bad
Obviously children shouldn't be exposed to any of this, but as long as the solution is either implementing a privacy violating surveillance scheme, or gatekeeping so that only the rich and powerful have unrestricted/uncensored access then I'm vehemently opposed to it and will gladly die on this hill.
Unlike most search engines or the internet itself, Facebook requires an account and asks for your age to use most of their features.
Where I wouldn't want to see the internet put behind some universal age verification, I would expect Facebook to act reasonably with regards to content children see given that they already know how old you claimed to be.
I have no idea how this maps to modern tech — just kind of pointing out that privacy and surveillance of course were not involved then.
The first easy step is to ban smartphones from schools, and that is happening in some places already, but the process is slow. Smartphones are a weapon of mass destruction for kids‘ attention spans.
Last but not least, parents need to be more conscious about their screen usage. I find it hard myself to resist the temptation of „checking the phone“, how can we expect kids to do so…
For research purposes, some availability of unrestricted AI is almost a necessity, but I don't think the same is true for the output of text generation on a (commercial) platform.
Careless?
Are we suddenly in favor of censorships?
This bot had access to the profile of its users and offered sexually explicit content to profiles that said the user was underage.
Your best case scenario, assuming you don't get laughed off the stage, is you addict a generation into "socializing" with robots.
I don't see it taking off, but like, obvious play.
This is pretty much, "I bought this hammer from Facebook. Look what happens when I use it to hit myself in the head." ... "OMG! Facebook's hammer hit me in the head! I'm shocked! Think of the children!"
The chatbot "knows" when they are talking to kids? Or are you saying kids would never ask such questions?
And just to head off the obvious response, yes, I guess a 17 year old is still technically a kid. But that's not what people are worried about here. Let's not conflate.