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

Losing Trust in Google

twitter.com

Post reply on HN
Print view View original

31–40 of 72 posts

Re: Losing Trust in Google

#31

Post by TomSwirly »

_Reo wrote: post #25

When you’re aware that Google’s “agenda” is just attempting to include more types of people by default in output, posts like this sound bizarre. Some of the scenarios Gemini generated were ridiculous, but so too are the stereotypes like “CEO = white male” that Stable Diffusion and other models would assume. It’s hard to trust that people this intensely upset about Google’s issues aren’t seeking to benefit from leavin…

Agree 100%.

The idea that "This can't be a bug! They must have spent huge amounts of effort making deliberately ridiculous outputs that no one would value!" comes close to clinical paranoia.

Re: Losing Trust in Google

#32

Post by rob74 »

pljung wrote: post #22

Did he also loose trust in OpenAI back when DALL-E exclusively generated images of white males when you prompted it with "CEO"? And images of women when prompted with "nurse"?[0] Or is this just the old google-bashing? [0] https://www.vice.com/en/article/wxdawn/the-ai-that-draws-wha...

Not sure if it's Google-bashing, or if this particular issue with Gemini hit the "anti-woke" nerve of some people, but I agree that this reaction is way overblown. Some AI models err on the side of reinforcing clichés, Google tried to avoid that and overdid it with some bizarre results - so what?

Re: Losing Trust in Google

#33

Post by gopher_space »

dns_snek wrote: post #12
fsh wrote: post #9

Earlier quoted context omitted.

Gemini tends to generate images of non-white people which made some white people very mad.

Take off the spin. When asked to generate nazi soldiers, it generated black, asian, and native american people wearing nazi uniforms. When asked to generate white people, it said it's not allowed to.

[flagged]

Re: Losing Trust in Google

#34

Post by anothernewdude »

fshbbdssbbgdd wrote: post #13

If you make a language model which just predicts the next word without trying to shape the output to be “good”, what you get is GPT instead of ChatGPT. You try to ask it a question and it will keep asking more questions (with a similar writing style). Or curse at you, or change the subject. RLHF was the breakthrough that fixed this. We used human feedback to train the model to give “good” output. Ok, so probably we a…

GPT was good. ChatGPT ruined it. It is better and more powerful to write prompts such that the continuation is what you want, than ask questions as if it understood.

Now we don't even get log probs, one of the most powerful features to evaluate the output of the models.

Re: Losing Trust in Google

#35

Post by beardyw »

Seems to me that Google is trying to deal with ingrained bias contained in all of the data it slurps up. I think this is an issue that all AI has to deal with. It is inevitable that content which AI produces will become part of the data AI uses to build its models. If no action is take that echo chamber will intensify those biases on every iteration.

It looks as if Google is trying to tackle the problem in a somewhat piecemeal way, and have got it wrong. But from an AI perspective its no worse than people with 8 fingers or whatever, just more shocking.

I would rather that those working on AI were aware of the issue and getting it wrong, than simply not caring at all. Our children don't deserve to live in a world even more distorted than the one we live in.

Re: Losing Trust in Google

#36

Post by dns_snek »

bjord wrote: post #15

99% sure this is the "google hates white people" thing that a specific set of people have been absolutely losing their minds about gemini produced images of non-white people in a lot of situations in which it shouldn't have I've read theorized(?) that, in order to counteract disproportionately large amounts of pictures of white people in training data, they basically added instructions after the fact in an effort to…

Other models have biases in outputs and conform to stereotypes, but when used as designed, as a creative tool, that's not a real issue as long as you can correct the output with a simple prompt adjustment.

The unique issue with Gemini is that it would flat out refuse to follow simple prompts such as "Please generate an image of a white family" because they "weren't diverse and inclusive" enough, but if you changed "white" to any other qualifier, it happily obliged.

Re: Losing Trust in Google

#37

Post by fshbbdssbbgdd »

anothernewdude wrote: post #34
fshbbdssbbgdd wrote: post #13

If you make a language model which just predicts the next word without trying to shape the output to be “good”, what you get is GPT instead of ChatGPT. You try to ask it a question and it will keep asking more questions (with a similar writing style). Or curse at you, or change the subject. RLHF was the breakthrough that fixed this. We used human feedback to train the model to give “good” output. Ok, so probably we a…

GPT was good. ChatGPT ruined it. It is better and more powerful to write prompts such that the continuation is what you want, than ask questions as if it understood. Now we don't even get log probs, one of the most powerful features to evaluate the output of the models.

Raw GPT is a good tool but there’s a reason it didn’t get 100 million users.

If you made unbiased Gemini and asked it for images of German soldiers it probably wouldn’t output an image at all, and if it did it would probably be porn.

Re: Losing Trust in Google

#38

Post by dns_snek »

gopher_space wrote: post #33
dns_snek wrote: post #12

Earlier quoted context omitted.

Take off the spin. When asked to generate nazi soldiers, it generated black, asian, and native american people wearing nazi uniforms. When asked to generate white people, it said it's not allowed to.

[flagged]

You're making unfounded assumptions, I'm neither of those things. I'm disappointed and extremely entertained by the whole ordeal.

Re: Losing Trust in Google

#39

Post by RugnirViking »

bjord wrote: post #30
bjord wrote: post #15

99% sure this is the "google hates white people" thing that a specific set of people have been absolutely losing their minds about gemini produced images of non-white people in a lot of situations in which it shouldn't have I've read theorized(?) that, in order to counteract disproportionately large amounts of pictures of white people in training data, they basically added instructions after the fact in an effort to…

grow up and explain why I'm wrong, instead of just downvoting me into oblivion—if I'm really so far off-base, wouldn't you want to correct my understanding of the situation?

personally im not bothered by the output exactly per se: it would be great if they modified the dataset to be more inclusive instead. The invisible prompt editing is what really gets me. The fact that the ai is instructed after the fact to make changes to what you asked it to do, invisibly. These things really, REALLY need to be tool-shaped, if they do all sorts of hidden extra bits they will be vastly less useful and leaves room for vastly more nefarious ends.

Like would it be so hard for it to just show the prompt it submits to dall-e? help people learn how the system works?

Re: Losing Trust in Google

#40

Post by aredox »

dns_snek wrote: post #36
bjord wrote: post #15

99% sure this is the "google hates white people" thing that a specific set of people have been absolutely losing their minds about gemini produced images of non-white people in a lot of situations in which it shouldn't have I've read theorized(?) that, in order to counteract disproportionately large amounts of pictures of white people in training data, they basically added instructions after the fact in an effort to…

Other models have biases in outputs and conform to stereotypes, but when used as designed, as a creative tool, that's not a real issue as long as you can correct the output with a simple prompt adjustment. The unique issue with Gemini is that it would flat out refuse to follow simple prompts such as "Please generate an image of a white family" because they "weren't diverse and inclusive" enough, but if you changed "w…

Hey, look at the first website returned in "normal" google for "Please generate an image of a white family": https://www.alamy.com/stock-photo/white-american-family.html...

And Bing? https://www.gettyimages.ch/fotos/white-family

OTOH, if you think that "Please generate an image of a white family" is good use of those behemoth language models and all the rsssources poured in...

Post reply on HN