New Mexico sued Snapchat after using AI to reveal child safety risks.
New Mexico is seeking an injunction to permanently block Snap from practices allegedly harming kids. That includes a halt on advertising Snapchat as "more private" or "less permanent" due to the alleged "core design problem" and "inherent danger" of Snap's disappearing messages. The state's complaint noted that the FBI has said that "Snapchat is the preferred app by criminals because its design features provide a false sense of security to the victim that their photos will disappear and not be screenshotted."
"Heather" also tested out Snapchat's search tool, finding that "even though she used no sexually explicit language, the algorithm must have determined that she was looking for CSAM" when she searched for other teen users.
But literally in just the previous paragraph:
Posing as "Sexy14Heather," the investigator swapped messages with adult accounts, including users who "sent inappropriate messages and explicit photos.
Gee, I wonder how the algorithm could've possibly suggested these users. What a mystery.
I'm not defending Snapchat here - they're a scumbag company with a scumbag product and they should be held responsible for enabling the sharing of CSAM on their platform - but it doesn't just match you with random predators out of thin air. They went in with specific keywords in their username and a pattern of account engagement.
That's nuance that I hadn't considered and I appreciate you pointing it out. I'm not on any of these sharing platforms so I have no idea what they're like and that made it easy to overlook this detail which is probably pretty relevant.
They're also trying to claim screen shots are a major security issue, even though everyone knows they can just do that and the real benefit of deleted messages is that no corpo, cop, or cracker can snoop on them later.
They are completely incompetent and tech illiterate, is what I'm saying
You know that on Snapchat the messages aren't deleted anymore right? They are stored on your phone and the Snapchat servers. That is how "memories" works, and there used to be screenshot workaround to grab the photo back after the fact, but I am not sure if they have hid them better by now.
Snapchat is not even close to a secure messaging program...
Despite Snapchat setting the fake minor's profile to private and the account not adding any followers, "Heather" was soon recommended widely to "dangerous accounts, including ones named 'child.rape' and 'pedo_lover10,' in addition to others that are even more explicit," the New Mexico DOJ said in a press release.
Tough call. If you put out bait, you’re gonna get someone. But would that person have done the same thing if they had not seen your bait? Chicken and the egg. On one hand, it looks like entrapment.
I mean, that part isn't really at issue here. It's fundamentally the same technique that's been used since the 90's, famously on To Catch a Predator. Seemingly, the "entrapment" angle has been settled.
But now they can argue that they aren’t sexually attracted to children, just AI artwork, which is technically not an image of a child. And unless I missed it, they were not trying to meet the girl.
The problem is going to be that images that aren’t real of a crime aren’t a crime. Of the opposite was true, images of murder would be illegal. Can’t just cherry pick.
If I draw a stick figure and label it “naked girl,” does it become child porn? What if I’m a really good artist?
I agree completely with you that Snapchat is an unmonitored disaster that gives the user the impression that the person they are sharing their nudes with cannot save the content. A good portion of the videos on porn sites have that little Snapchat progress wheel on them and are clearly screencaps.
Aside: I think there is a big topic that conveniently gets overlooked because it’s so much easier to blame “social media” or the “predator,” and that is—where are the fucking parents?
IANAL, but my understanding is entrapment is when they convince you to do something you might not otherwise have done. So if the cops create an account of a minor and message an adult asking if they want to fuck, and the answer is like "uh no, absolutely not," and then the cops follow up by repeatedly sexting, and the adult blocks their account, but the cops relentlessly keep sexting from burner accounts, and plant people in the adult's work and social environments who keep talking about how normal it is to fuck minors who sext you out of the blue, and then the adult is finally like "oh whatever, fine" - that's entrapment.
Now, most people still are literally never going to take the minor up on the offer, no matter how relentless they are or how normalized it is in their environment. That's true about most crimes. The question is how many people wouldn't have committed that crime unless this very specific police-created situation came up, and that difference is what falls into entrapment.
I'd argue this isn't even close to entrapment, because all they did was set up an account much like all the others that exist, and waited for others to find them. It's no different from leaving a bike unlocked, then catching somebody who steals it. There are unlocked bikes everywhere, and people don't suddenly decide to steal the only bike of their life because they happened to find that unlocked bike.
Of course, they could also be spending this time and money getting to the root of societal issues and fixing the core problems instead of catching a small percentage of active pedophiles and letting the rest of them continue to cause irreparable harm.
The last paragraph is the big issue. Fix society. We can argue all day long about what line does “artwork” cross before it becomes illegal, but that’s not actually preventing anyone from getting abused.
And imo, it seems a little sick to say, “we made a bunch of kiddie porn that didn’t exist previously, and I’m going to distribute it to catch criminals—using tax dollars” … tf?
In general I think using AI imagery, and catfishing in general, is basically entrapment. In most civilized countries; that's illegal for police to do.
Now if they begin to actually trade in actually legitimate forbidden materials...sure; by all means arrest for that charge alone. That wouldn't be unjustified. But provoking someone who might then turn around and harm a real child, seems wrong.
Maybe we should just make those AI tools public and free and leave it at that. If you monsters want kiddie porn, have this AI generated stuff. It's sort of like having needle exchanges or safe using centers for addicts. It doesn't solve the problem, but it makes it safer for everyone.
If we're talking about it from a harm reduction or public health view point then it is the circulation of the content, not the personal generation that poses the greatest risk. Once the material gets circulated, it becomes harder to know it's origins and it might look like someone IRL and expose them to risks.
I don’t understand how this would be fine but pedophiles generating them at home without distributing them would be illegal.
Cause the cops are creating CSAM and putting it out there in some shape or form, which you could argue would encourage pedophiles same way circulating ai images would as well.
Either generating under age sensitive material is always wrong or it’s not.
is, well, yeah . Like any major network you have creeps, Nazis and pedos and it's good to root them out. It's weird that themselves don't do that.
is ... difficult. I feel like the cops that prompt GenAIs to produce (sexualized) images of children risk severe trauma. It must be punishing to ask a a machine to produce fake images of morally reprehensible material. That's gotta take a toll on you. What's even more important is that those people could be spending their time taking down actual child abuse images.