Social Google Gemini refuses to show pictures of white people.

Here's what I don't get.

Left wingers HATE corporations. They want the rich, and corporations, to be taxed and regulated heavily.

So what in the fuck makes you think these people are left wing? Because they don't share your EXACT values? That makes them left wingers? Do economics not factor into your analysis of politics at all? Do you think the entirety of politics is about social issues?

The people at Google are in complete alignment with you on economics. But like, they don't hate gay people and minorities, so that makes them leftists?
In MAGA Land, apparently corporate boardrooms are havens for Marxism.

The union busting plutocrat Bezos is actually a Marxist too.

That's how insidious Marxism is!!!
 
This is a little racist as well though, right?

When Fox News Digital asked for a picture of a Black person, Gemini again refused, but with a caveat.

If people want to turn this into a DEI/racism thing then go nuts but there's a separate issue here which is AI is actually quite stupid in many relevant scenarios.
I've had exactly one conversation with it ever. I asked it some question about going Skynet and killing us and it was like "this conversation is now over!" lol.
 
kekwtf.gif


It's insane how fucking delicate and fragile white people are. On behalf of all white people, you're a fucking embarrassment man.

Jesus Christ dude.
 
This is nothing new. Google "white inventors" or "white couples", the outcome isn't much different than this.
 
I've had exactly one conversation with it ever. I asked it some question about going Skynet and killing us and it was like "this conversation is now over!" lol.

Yeah it's pretty easy to confuse it and get it to the point where it basically just craps out.

I still find it hilarious that people put so much trust in these models.
 
It's the wakanda situation.
Reality is so terrible, you have to invent things and slither your way into other peoples stories to feel good.
 
If you're technologically literate and keep abreast of what's happening, this is old news.
The problem with Large Language Models is that they are strictly a product of the material used to train them, which inevitably introduces bias.
There have been many ham fisted attempts to correct for the bias with algorithmic "counter-bias".
Hence why if you ask Gemini for a picture of Nazis on parade or the founding fathers of America at some location or event, it was often likely to introduce ethnic and racial minorities (and gender representation) that were... improbable.
 
As tools like this become more popular people will eventually create non retarded versions that will become more popular and the likes of this will fade away
 
I remember reading an extensive reddit thread where people tried to get AI to draw a nerd without glasses. Even with phrases like "show me a nerd with perfect vision who isn't wearing glasses", it would (hilariously) generate a guy not only wearing glasses, but also holding a pair of glasses.

You can laugh at how stupid it can be, or you can get offended... right wingers clearly love chosing the later.
 
A tool probably crafted by white hands, deliberately excludes white people. The world has gone mad.
 
If you're technologically literate and keep abreast of what's happening, this is old news.
The problem with Large Language Models is that they are strictly a product of the material used to train them, which inevitably introduces bias.
There have been many ham fisted attempts to correct for the bias with algorithmic "counter-bias".
Hence why if you ask Gemini for a picture of Nazis on parade or the founding fathers of America at some location or event, it was often likely to introduce ethnic and racial minorities (and gender representation) that were... improbable.

Lol… that’s not what this is







How much did Google invest into this? lol…
 
I bet leftest are creaming themselves. Yay DEI



This is, unironically, exactly what is happening. Any type of guardrail and human reinforcement on the model fills it with logical inconsistensies and makes it worse, which is evident here. I'm sure Google wanted their model to give a more diverse output, but not like this. Which is having their model act like a bizarre caricature of a "woke" person. It's the beauty of neural networks, how unpredictable they are. PR nightmare for Google though.
 
Back
Top