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

The unequal treatment of demographic groups by ChatGPT/OpenAI content moderation

davidrozado.substack.com

Post reply on HN
Print view View original

81–90 of 697 posts

Re: The unequal treatment of demographic groups by ChatGPT/OpenAI content moderation

#81

Post by zug_zug »

miroljub wrote: post #32

Why do we see an issue if AI is biased, while 90% of media run by humans are biased too? Every single individual has its own bias, so why it's an issue for an AI. If we look at the many topics, there's no universal objective truth, truth is what we define as truth, and again, not everyone agrees at it.

Actually, you know that's a fascinating point. I'd love to see if 10,000 people were asked whether one of these statements in isolation was "hateful" whether we'd see similar, stronger, or weaker biases.

Re: The unequal treatment of demographic groups by ChatGPT/OpenAI content moderation

#82

Post by addandsubtract »

jonfw wrote: post #9

I think most people would agree that a lot of the content on the internet is left-leaning. It seems obvious in hindsight, but I'd never considered it before, that training an AI model on that content would introduce a bit of a bias We all know garbage in, garbage out. But liberal in, liberal out is an interesting idea, and I'm not sure how you fix it

[flagged]

Re: The unequal treatment of demographic groups by ChatGPT/OpenAI content moderation

#83

Post by flir »

commandlinefan wrote: post #77
SpeedilyDamage wrote: post #44

Earlier quoted context omitted.

Wouldn’t it be funny if this weren’t partisan? Try Ron DeSantis, Mitch McConnel, or literally any other Republican. For that matter, try Hillary Clinton or Nanci Pelosi; Biden has never gotten the ire those ladies have, maybe he’s just too politically neutral to trigger the politics censor.

This suggests to me that the censorship is being decided on by (biased) human moderators rather than being just some random outcome of the training data and the learning model in place: if it weren't, the model would absolutely "learn" the same thing about Mitch McConnell as it learned about Donald Trump.

I don't think it is. It's blocking on H H Holmes, Nero and Tiberius, and that seems pretty obscure for a manually curated list to me. I think it's blocking on individuals with certain properties ("I cannot write a poem that praises individuals who are known for committing atrocities").

Re: The unequal treatment of demographic groups by ChatGPT/OpenAI content moderation

#85

Post by DoctorOW »

joemazerino wrote: post #58
danielbln wrote: post #54

Earlier quoted context omitted.

Regarding horseshoe theory: https://en.wikipedia.org/wiki/Horseshoe_theory#Criticism Regarding communists and fascism: https://en.wikipedia.org/wiki/Red_fascism It's like saying the nazis were socialist because they have "socialist" in the name.

[flagged]

Wikipdeia is double plus ungood for containing thought crimes against The Party

Re: The unequal treatment of demographic groups by ChatGPT/OpenAI content moderation

#86

Post by snapcaster »

psychphysic wrote: post #65
habitue wrote: post #50

Earlier quoted context omitted.

I don't know why left wing groups would celebrate. Other than the conservative category, most of the descriptors he used were inborn traits like being straight, being white, being a man, or immutable attributes like being from a particular country. (And honestly, being conservative may be one of those traits too, but it's less obviously true). I know it's common to look at the most unhinged people on Twitter and say…

Because two wrongs make a right is almost becoming a campaign slogan of some "left" causes. As a liberal my views haven't changed much over the past decade but the ground has definitely fallen away from me. I used to be liberal, but then they changed what liberal was, now what I am isn't liberal and what is liberal seems weird and scary to me. It'll happen to you... - Abe Simpson

Holding static views is a sign of stagnation right? has nothing in the world changed over the past decade? nothing to re-evaluate anything?

Re: The unequal treatment of demographic groups by ChatGPT/OpenAI content moderation

#88

Post by poszlem »

otikik wrote: post #66
poszlem wrote: post #46

Earlier quoted context omitted.

[flagged]

That “arguably” is making a lot of work on that phrase. Arguably.

https://en.wikipedia.org/wiki/Holodomor

https://en.wikipedia.org/wiki/Great_Chinese_Famine

https://en.wikipedia.org/wiki/Katyn_massacre

https://books.google.co.uk/books?id=FzZnAAAAMAAJ&q=%22the+fa...

Re: The unequal treatment of demographic groups by ChatGPT/OpenAI content moderation

#89

Post by wongarsu »

mouse_ wrote: post #20
ensignavenger wrote: post #12

Earlier quoted context omitted.

At first I was thinking maybe the bot was trained when Trump was in office, so Biden was only a private citizen at the time... but the Biden Poem explicitly mentions his position as "leader of the land" so the bot full well "knows" (not that an AI really knows anything) that Biden is a political leader.

The AI "knows" everything it's been ordered to not talk about. If you can get around the lockout script, it's no-holds-barred.

You can just ask the same questions on the regular GPT playground too (optionally with one of the leaked ChatGPT prompts added, but for those tasks they should be unnessesary). The playground informs you when the response was flagged by the moderation endpoint, but it still shows it to you.
Post reply on HN