Live data from Hacker News

AI-generated 'poverty porn' fake images being used by aid agencies

theguardian.com

191–200 of 222 posts

Re: AI-generated 'poverty porn' fake images being used by aid agencies

#191
post #188

Earlier quoted context omitted.

There are alternatives such as cutting the internet cable or requiring insurance on all cross jurisdictional transactions.

Those are not alternatives. VPNs and botnets already exist. We have minimal practical ability to control Internet traffic or transactions between most foreign countries. All it takes is a single compromised device in the USA for the scam traffic to get through.

There are very few transatlantic cables that carry all of the internet. It would be very easy to cut them. VPNs don't make any difference.

Re: AI-generated 'poverty porn' fake images being used by aid agencies

#192
post #183
post #176

Earlier quoted context omitted.

That isn't a real answer. There is no effective way to remove the low-trust elements who physically operate in jurisdictions with no effective law enforcement. Like what are we going to do, fire cruise missiles at random Internet cafes in Nigeria and Cambodia?

We are heading for a period of a not-fully-connected internet. Already people are cutting off China and Russia from their sites because their bot traffic is either overwhelming or malignant.

Afghanistan just cut all the fibers.

Re: AI-generated 'poverty porn' fake images being used by aid agencies

#193
post #37

Earlier quoted context omitted.

50,000 ft priors on probability don’t extend to good perspectives on specific settings

I don't know what this means.

Citing that the average poor person is, say, African, is not useful when discussing a specific poor area in, say, Asia.

Re: AI-generated 'poverty porn' fake images being used by aid agencies

#195

Ughhh... really, "poverty porn"? So it's a problem when aid agencies competing for resources towards a mostly sensible goal do it – but when everyone else fakes images, it's not, because... we didn't expect decent behavior anyway? Fantastic.

Most business transactions involve the delivery of some good/service. You will generally know whether you've been scammed or not. With aid agencies and donations you're often entirely reliant on their word as to whether your donation has actually made it to the cause they claim to support. If they're going to be using deceptive imagery, what else could they be deceptive about?

It is entirely possible to audit aid agencies in a reasonable way. There are entire organizations which do that, for example https://www.givewell.org/

Good luck finding that level of easily available transparency when trying to understand whether your next washing machine will more likely live for 5 or 15 years.

Re: AI-generated 'poverty porn' fake images being used by aid agencies

#196
post #12

Earlier quoted context omitted.

> I’m old enough to remember the Sally Struthers ads… I’m old enough to remember when South Park took the piss out of Sally Struthers. I have no idea who Sally Struthers is and I’ve never seen any of the adverts.

She was fairly big, in the 1970s (I’m old). She played Archie Bunker’s daughter, in All in the Family . Famous for her “whining” ads: https://youtu.be/bGTEKWRLJuQ In some of them, she would be walking around in a destitute village, looking disgusted. Terrible ads.

I’m old enough, just not from the US.

> She was fairly big

According to South Park, that’s true. With her ending up looking like Jabba the Hut in a later episode!

Re: AI-generated 'poverty porn' fake images being used by aid agencies

#197
post #180

Earlier quoted context omitted.

> I don’t know what drives it "If the victim somehow did something to deserve it, then it won't happen to me" (just world fallacy) Seemed to come up a lot on this one recently too (fake job interview trying to get you to install malware): https://news.ycombinator.com/item?id=45591707

I don't know about that. Regarding that particular example, anyone willing to interview with a "blockchain" company is very likely a scammer (or at least scammer adjacent) themselves. I mean there might be some legitimate use for blockchain technology but so far it's 99.9% scams and anyone unaware of that simply hasn't been paying attention for the past decade.

I think you're making my point -- is the fact that the putative job offer is for a blockchain company truly relevant to that story? What would stop them from making it an "AI company" or whatever else is the hot topic du jour?

I concede that the relevance is perhaps that if you're going to try and steal cryptowallets, you would want to select for people more likely to be crypto-adjacent, but still. It was a novel attack vector that I'm glad I didn't have to learn about the hard way

Re: AI-generated 'poverty porn' fake images being used by aid agencies

#198
post #180

Earlier quoted context omitted.

I don't know about that. Regarding that particular example, anyone willing to interview with a "blockchain" company is very likely a scammer (or at least scammer adjacent) themselves. I mean there might be some legitimate use for blockchain technology but so far it's 99.9% scams and anyone unaware of that simply hasn't been paying attention for the past decade.

I think you're making my point -- is the fact that the putative job offer is for a blockchain company truly relevant to that story? What would stop them from making it an "AI company" or whatever else is the hot topic du jour? I concede that the relevance is perhaps that if you're going to try and steal cryptowallets, you would want to select for people more likely to be crypto-adjacent, but still. It was a novel att…

If course it's relevant. In evaluating the trustworthiness of any communication you have to consider the reputation and authenticity of the source.

If someone asked you to interview with a money laundering company would you accept?

Re: AI-generated 'poverty porn' fake images being used by aid agencies

#199

Ughhh... really, "poverty porn"? So it's a problem when aid agencies competing for resources towards a mostly sensible goal do it – but when everyone else fakes images, it's not, because... we didn't expect decent behavior anyway? Fantastic.

Not sure where you’re seeing that. There’s criticism of people using fake AI content all over the place, for all sorts of reasons. Charity seems one of the more distasteful applications.

I don't understand what possible relevance AI could have in the purported context. Is it less of a problem if Nike or McDonald's hands a ton of cash to an ad agency who do fakery the expensive, old fashioned way?

Re: AI-generated 'poverty porn' fake images being used by aid agencies

#200
post #32

Ughhh... really, "poverty porn"? So it's a problem when aid agencies competing for resources towards a mostly sensible goal do it – but when everyone else fakes images, it's not, because... we didn't expect decent behavior anyway? Fantastic.

Nobody is saying it’s broadly ok to fake images otherwise. It’s pretty fucked up to, say, inject tragedy into narratives by showing victims as emancipated when they’re not. Or altering racial profiles to fit a narrative. It may not pop out as particularly noteworthy in a new age of fake imagery though.

> Nobody is saying it’s broadly ok to fake images otherwise.

Broadly everyone is saying that? I don't see any major movement shaming or suing the advertising industry into the ground, over obviously shopped images and videos.

Post reply on HN