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

AI behavior guardrails should be public

twitter.com

Post reply on HN
Print view View original

101–110 of 350 posts

Re: AI behavior guardrails should be public

#101

Post by ceejayoz »

DebtDeflation wrote: post #90
dekhn wrote: post #12

I strongly suspect Google tried really, really hard here to overcome the criticism is got with previous image recognition models saying that black people looked like gorillas. I am not really sure what I would want out of an image generation system, but I think Google's system probably went too far in trying to incorporate diversity in image generation.

Surely there is a middle ground. "Generate a scene of a group of friends enjoying lunch in the park." -> Totally expect racial and gender diversity in the output. "Generate a scene of 17th century kings of Scotland playing golf." -> The result should not be a bunch of black men and Asian women dressed up as Scottish kings, it should be a bunch of white guys.

You can see how this gets challenging, though, right?

If you train your model to prioritize real photos (as they're often more accurate representations than artistic ones), you might wind up with Denzel Washington as the archetype; https://en.wikipedia.org/wiki/The_Tragedy_of_Macbeth_(2021_f....

There's a vast gap between human understanding and what LLMs "understand".

Re: AI behavior guardrails should be public

#102

Post by sho_hn »

michaelt wrote: post #51
dekhn wrote: post #12

I strongly suspect Google tried really, really hard here to overcome the criticism is got with previous image recognition models saying that black people looked like gorillas. I am not really sure what I would want out of an image generation system, but I think Google's system probably went too far in trying to incorporate diversity in image generation.

As well as that, I suspect the major AI companies are fearful of generating images of real people - presumably not wanting to be involved with people generating fake images of "Donald Trump rescuing wildfire victims" or "Donald Trump fighting cops". Their efforts to add diversity would have been a lot more subtle if, when you asked for images of "British Politician" the images were recognisably Rishi Sunak, Liz Truss…

My takeaway from all of this is that alignment tech is currently quite primitive and relies on very heavy-handed band-aids.

Re: AI behavior guardrails should be public

#103

Post by nostromo »

matt3210 wrote: post #58

How is this any different than doing google image searches of the same prompts. Exmaple: Google image search for "Software Developer" and you get results such that there will be the same amount of women and men event though men make up the large majority of software developers. Had Google not done this with its AI I would be surprised. There's really no problem with the above... If I want male developers in image sea…

Yes, Google has been gaslighting the internet for at least a decade now.

I think Gemini has just made it blatantly obvious.

Re: AI behavior guardrails should be public

#104

Post by AlecSchueler »

DebtDeflation wrote: post #90
dekhn wrote: post #12

I strongly suspect Google tried really, really hard here to overcome the criticism is got with previous image recognition models saying that black people looked like gorillas. I am not really sure what I would want out of an image generation system, but I think Google's system probably went too far in trying to incorporate diversity in image generation.

Surely there is a middle ground. "Generate a scene of a group of friends enjoying lunch in the park." -> Totally expect racial and gender diversity in the output. "Generate a scene of 17th century kings of Scotland playing golf." -> The result should not be a bunch of black men and Asian women dressed up as Scottish kings, it should be a bunch of white guys.

As soon as you have then playing an anachronistic sport you should expect other anachronistic imagery to creep in, to be fair.

Re: AI behavior guardrails should be public

#105

Post by AlecSchueler »

_heimdall wrote: post #87
redox99 wrote: post #66

Earlier quoted context omitted.

I remember checking like a year ago and they still had the word "gorilla" blacklisted (i.e. it never returns anything even if you have gorilla images).

Gotta love such a high quality fix. When your upper high tech, state of the art algorithm learns racist patterns just blocklist the word and move on. Don't worry about why it learned such patterns in the first place.

[flagged]

Re: AI behavior guardrails should be public

#106

Post by ceejayoz »

AlecSchueler wrote: post #104
DebtDeflation wrote: post #90

Earlier quoted context omitted.

Surely there is a middle ground. "Generate a scene of a group of friends enjoying lunch in the park." -> Totally expect racial and gender diversity in the output. "Generate a scene of 17th century kings of Scotland playing golf." -> The result should not be a bunch of black men and Asian women dressed up as Scottish kings, it should be a bunch of white guys.

As soon as you have then playing an anachronistic sport you should expect other anachronistic imagery to creep in, to be fair.

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

> The modern game of golf originated in 15th century Scotland.

Re: AI behavior guardrails should be public

#107

Post by alpaca128 »

AlecSchueler wrote: post #105
_heimdall wrote: post #87

Earlier quoted context omitted.

Gotta love such a high quality fix. When your upper high tech, state of the art algorithm learns racist patterns just blocklist the word and move on. Don't worry about why it learned such patterns in the first place.

[flagged]

It's not unavoidable, but it would cost more to produce high quality training data.

Re: AI behavior guardrails should be public

#108

Post by mike_d »

Jason_Protell wrote: post #47
stainablesteel wrote: post #40

gemini seems to have problems generating white people and honestly this just opens the door for things that are even more racist [1], the harder you try the more you'll fail, just get over the DEI nonsense already 1. https://twitter.com/wagieeacc/status/1760371304425762940

Is there any evidence that this is a consequence of DEI rather than a deeper technical issue?

It is possible Google tried to avoid likenesses of well known people by removing any image from the training data that contained a face and then including a controlled set of people images.

If you give a contractor a project that you want 200k images of people who are not famous, they will send teams to regions where you may only have to pay each person a few dollars to be photographed. Likely SE Asia and Africa.

Re: AI behavior guardrails should be public

#109

Post by CuriouslyC »

photoGrant wrote: post #54
dekhn wrote: post #12

I strongly suspect Google tried really, really hard here to overcome the criticism is got with previous image recognition models saying that black people looked like gorillas. I am not really sure what I would want out of an image generation system, but I think Google's system probably went too far in trying to incorporate diversity in image generation.

Remind yourself we're discussing censorship, misinformation, inability to define or source truth and we're concerned on Day 1 about the results of image gen being controlled by a for profit single entity with incentives that focus solely on business and not humanity... Where do we go from here? Things will magically get better on their own? Businesses will align with humanity and morals, not their investors? This is…

The ridiculous degree of PC alignment of corporate models is the thing that's going to let open source win. Few people use bing/dall-e, but if OpenAI had made dall-e more available and hadn't put ridiculous guardrails on it, stable diffusion would be a footnote at this point. Instead, dall-e is a joke and people who make art use stable diffusion, with casuals who just want some pretty looking pictures using midjourney.

Re: AI behavior guardrails should be public

#110

Post by whatwhaaaaat »

tobbe2064 wrote: post #85
goatlover wrote: post #62

Earlier quoted context omitted.

You want large tech companies "creating reality" on behalf of everyone else? They're not even democratic institutions that we vote on. You trust they will get it right? Our benevolent super rich overlords.

Its not really a question about want, its a question about facts. Their actions will make a significant mark on the future. So far it seems like they are trying to promote positive changes such as inclusion and equality. Which is far far far fucking really infinitely far better than trying to promote exclusion and inequality

You are so right! Just not the way you want to be.

Google and the rest of “techs” ham fisted approach has opened the eyes of millions to the bigotry these companies are forcing on everyone in the name of “improvement” as you put it.

Post reply on HN