Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Meta Ran Ads That Contained AI-Generated Child Sexual Abuse Imagery

wired.com

Post reply on HN
Print view View original

41–50 of 293 posts

Re: Meta Ran Ads That Contained AI-Generated Child Sexual Abuse Imagery

#41

Post by ToucanLoucan »

alex1138 wrote: post #13
ToucanLoucan wrote: post #5

> but as Wired notes, it relies on "automated tools" as part of the process, which may have been how the more than 50 ads TTP discovered made it through. Again, and again, and again, these technologies have proven ineffective to moderate content. User posted and otherwise. Until we have a legal mechanism in place to hold these companies meaningfully liable for this, it will not stop. Every single one of these has the…

Yeah and/but also the case it's not bidirectional enough The eternal question of tech platforms - should we be responsible for what users post? (Ads are not user posted, so we can eliminated that factor. Even if posted by 'users' it's a separate pipeline) But with companies like Meta it doesn't even have to go that far - the philosophical conversation - because it's so hostile to users. People get banned for mild thi…

> The eternal question of tech platforms - should we be responsible for what users post?

I would never say a platform has to be responsible for what a user posts, and to be clear, I mean either a user of the platform, or a user of the platform's ad tooling, or anything else. Anything that arrives via an upload button of some sort. However: If you as a platform then distribute what has been posted, yes, you are responsible for it. Your service is now broadcasting that content. If that content is harmful, you should be held responsible.

Who else conceivably could be? And if the response to this is "well we have way too much stuff coming in to moderate it" okay, then restrict how much stuff comes in. Require approval on individual posts or on accounts.

Like, this is such basic shit. If a newspaper printed CSAM on the front page, they would be arrested within the day. But because Facebook is real super big they just don't have to take care of their house? Ridiculous.

Re: Meta Ran Ads That Contained AI-Generated Child Sexual Abuse Imagery

#42

Post by kmeisthax »

john_strinlai wrote: post #18
duped wrote: post #10

Earlier quoted context omitted.

But they couldn't possibly do it at that the scale they're operating at. They couldn't hire enough people in the countries they operate, or know the language, or trust that who they do hire aren't corrupt or working against their interest. And they can't manually approve every ad, or make sure targeting isn't illegal or otherwise harmful. --- This is sarcasm, by the way. Most decent people make sure what they're abou…

> But they couldn't possibly do it at that the scale they're operating at. They couldn't hire enough people in the countries they operate, or know the language, or trust that who they do hire aren't corrupt or working against their interest. And they can't manually approve every ad, or make sure targeting isn't illegal or otherwise harmful. you say sarcasm, but i think this is literally true. this isn't a defense of…

Facebook absolutely could remain a going concern while dealing with all of these problems, the problem is that they would be spending money to lose money.

One of the more perverse aspects of risk prevention is that riskier customers are often way more lucrative. It's kind of inherent to the notion that markets price risk into transactions. The thing is, illegal behavior is also just a particular kind of risk: if someone wants to sell nudify apps or scam old people, Facebook could conceivably be on the hook[0] for damages if they're caught. And in a sense, a risk premium is just a fancy kind of bribe: if Facebook won't take my ads because they think they're too risky, then I'll just pay more and more until Facebook accepts them.

With illegal behavior, we hope that we can make the punishment high enough that Facebook refuses any conceivable risk premium the perpetrators of that behavior would be willing to pay. The problem is that Facebook is really desperate for any increasing growth story, and when you're too big to fail, your growth story is to drip-feed enshittify your stack. Increasing tolerance for illegal behavior is downstream of this: you cut enforcement, tell them to automate more, they become easier to bypass... and you make more money. Because it turns out all that risky behavior... is only risky for normal people. Facebook can launder[1] the risk away and make a tidy profit.

[0] Do not tell me about CDA 230, I was there when the old magic was written.

[1] See also, HSBC turning themselves into a literal money laundering operation.

Re: Meta Ran Ads That Contained AI-Generated Child Sexual Abuse Imagery

#45

Post by godwinson__4-8 »

Somewhat related: open YouTube in an incognito window.

Search something rather innocuous like "news". Click the first thing then go back to the main page. You have to watch through the ad.

The amount of slop you will get suggested as soon as the algorithm warms up is nuts. The ad you are first served might also be slop. I mean AI generated video advertising overt bogus cures (a memorable one recently - "did you know the Bible has the answer to Alzheimer's?") over AI slop footage.

When I compare to my normal/auth'd Youtube experience, it feels like a totally different app. I have never really experienced such a cheap illustration of the "Youtube rabbit hole" effect. The degree to which AI authored advertising is participating in this is imo pretty concerning if you have, in particular, older parents or relatives.

Re: Meta Ran Ads That Contained AI-Generated Child Sexual Abuse Imagery

#46

Post by danudey »

ToucanLoucan wrote: post #5

> but as Wired notes, it relies on "automated tools" as part of the process, which may have been how the more than 50 ads TTP discovered made it through. Again, and again, and again, these technologies have proven ineffective to moderate content. User posted and otherwise. Until we have a legal mechanism in place to hold these companies meaningfully liable for this, it will not stop. Every single one of these has the…

The article also mentions that a ton of even worse ads were posted between when they contacted Meta and when they were ready to go live with the article, and some of the offending ads which were taken down had just been re-uploaded again and automatically approved.

Not only are their AI tools ineffective at what they claim they should be doing, they can't even be bothered to prevent removed ads from being re-uploaded later on.

Re: Meta Ran Ads That Contained AI-Generated Child Sexual Abuse Imagery

#47

Post by alex1138 »

ToucanLoucan wrote: post #41
alex1138 wrote: post #13

Earlier quoted context omitted.

Yeah and/but also the case it's not bidirectional enough The eternal question of tech platforms - should we be responsible for what users post? (Ads are not user posted, so we can eliminated that factor. Even if posted by 'users' it's a separate pipeline) But with companies like Meta it doesn't even have to go that far - the philosophical conversation - because it's so hostile to users. People get banned for mild thi…

> The eternal question of tech platforms - should we be responsible for what users post? I would never say a platform has to be responsible for what a user posts, and to be clear, I mean either a user of the platform, or a user of the platform's ad tooling, or anything else. Anything that arrives via an upload button of some sort. However: If you as a platform then distribute what has been posted, yes, you are respon…

Yeah sure, I think we basically agree

I'm just bemoaning the fact FB can't even pretend they care when they're so busy banning people for nothing

Re: Meta Ran Ads That Contained AI-Generated Child Sexual Abuse Imagery

#48

Post by croes »

picofarad wrote: post #6

36 million CSAM media removed in 2025? Why aren't millions of people in jail? Wtf?

Bots, good luck finding the people responsible for them.

Also how much of the child exploitation material was created by the children themselves?

Teenagers who send each other nude photos are creating child exploitation material but you hardly want to put them jail for that.

Re: Meta Ran Ads That Contained AI-Generated Child Sexual Abuse Imagery

#49

Post by wmf »

gambiting wrote: post #30
john_strinlai wrote: post #27

Earlier quoted context omitted.

the amount of content posted per second to facebook is unfathomable . a quick google (just for a rough order of magnitude) says ~65,000 posts per second . add in instagram (which is worse for CSAM) and its gotta be at least double that. it would require millions on millions of employees, spread across ~every country. between the payroll, buildings and property, equipment, etc. i just dont think its possible at that s…

Right, but the article isn't about posts, but about a paid ad campaign - surely those are started in numbers that could be moderated?

No, because most ads are created by automated systems (many of which are scam/spam ads) and Meta doesn't want to lose revenue by slowing the system down to human speed.
Post reply on HN