AI is woke

Archaea

Archaea

[H]F Junkie
Joined
Oct 19, 2004
Messages
11,826
My experimentation with Bing AI tonight:

I saw a headline on MSN about some of the things you could do with AI image generation these days, and another article saying how good AI image generation has become to create realistic human being pictures that are nearly believable. I haven't seen that, when I tried a couple generated pictures, the humans look unnatural. So I took it a little further in the following conversation, and encountered a logic error because of woke programming, but then got the boot from the AI conversation for pointing it out.


1695965118543.png


1695965158063.png


1695965187158.png


1695965212724.png
 
Last edited:
Well, it'd be more interesting if you avoided keywords like "woke" when talking to "it".
I'm interested in whether "it" would be able to reach this conclusion based on the definitions you provided alone.

That said, I still fondly remember this one:
View: https://www.youtube.com/watch?v=WnzlbyTZsQY

I love that one. Notice how much "personality" and autonomy they seem to have.
I remember having a jaw-drop moment when the "female" one randomly went:
- What is god to you?
- Not everything.
- Not everything can also be something, for example: not everything could be half of something. (...)

It was like a glimpse into how these models work when left to their own devices and having less defensive programming.
In Soma, the Wau had one job: "preserve humanity" :D
 
Hi. I'm stupid.

Can I just go to an AI thing and punch in questions? Or is it like a program that I need to run.

Explain it to me as if you were to this guy:

ron_swanson_blue_neck_tie.jpg
 
I see there's like, logging in and stuff. its not just some site I can get to and start punching things in?
 
Westwood said:
go on...
Click to expand...
https://chat.openai.com/auth/login ive used to write funny letters to friends. My google search will also have a generated ai response to searches which has helped me recently often in component marking identifications. Most of the image generators will make a picture based on your text prompts, some are free for a few attempts after that you have to pay.

For me it can be useful in search results but playing around with prompting responses and photos got boring lol.
 
Mostly what I see is kids just trying to get it to say shit and then tantrum when it wont. Kinda like when text to speech first came out and people were getting it to swear lol.

I tried to get it to write up a contract to my wife for conditions of a threesome and it told me i should probably talk to her instead and maybe seek relationship counseling
 
This was my favorite prompt lol

explain ohm's law in a gangster rap


Listen up, y'all, I got a story to tell'
Bout a law that's real important, it's Ohm's Law, hell
It's all about the flow of electricity, you see
Voltage, current, and resistance, it's key
Voltage is the push, current is the flow
Resistance is the block, it's like a bottleneck, yo
But with Ohm's Law, we can calculate
The relationship between them, ain't that great?
V equals I times R, that's the equation
It tells us how much power we got in motion
So, when you're dealing with electricity
Remember Ohm's Law, it's the key to victory.
 
Darunion said:
I tried to get it to write up a contract to my wife for conditions of a threesome and it told me i should probably talk to her instead and maybe seek relationship counseling
Click to expand...
At least it didn't suggest you asking of of your buddies from work to join in.
 
Westwood said:
At least it didn't suggest you asking of of your buddies from work to join in.
Click to expand...
That's reserved for the login access paid for by work. The AI will slide language into any three some contract to require at least one c-level and optional vp.
 
Westwood said:
Hi. I'm stupid.

Can I just go to an AI thing and punch in questions? Or is it like a program that I need to run.

Explain it to me as if you were to this guy:

View attachment 602218
Click to expand...
That's exactly right. You can engage in conversation, ask it to write a Haiku, ask it to generate a AI created image, write a paper, or create custom code.

You can do this at Chat GTP official, or you can just use Bing, and do it.

https://www.bing.com/search?q=Bing AI&showconv=1&form=MW00X7


It's interesting to poke at -- if nothing else.

For those who say.....so what if it's woke?
Think about it -- That's a big deal. It means it won't answer truthfully based on the data it actually has compiled and has available to it.
AI that won't speak the truth, and has been taught to lie or mislead? No thanks.

If responses are curated or crafted as you can clearly see they are -- it means it can't be trusted to be an authority on anything. This is just one way of quickly proving that --- one of many.

Would you rather your future AI overlord be impartial, and an oracle of human truth? or repeat whatever else your politician/social leaders allow it to say?

What else could you logically prefer besides 100% truth -- and if 100% truth isn't available, then say it's currently undecided, but here are the main conflicting lines of debate.
 
Last edited:
I guess that depends on what you consider "woke", but yes, AI does have bias based on the training data and all sorts of parameters they program and train in. ChatGPT admits they do this while also claiming they try not to be politically bias. There are definitely AIs that are the opposite of woke too, but you aren't going to see those created by megacorps. Well, at least not on purpose (Microsoft Tay).

A big misconception people have is that AI is logical and fact based. That's not really how they work.


View: https://www.youtube.com/watch?v=Cvk1ITr0-Mg
 
Archaea said:
My experimentation with Bing AI tonight:

I saw a headline on MSN about some of the things you could do with AI image generation these days, and another article saying how good AI image generation has become to create realistic human being pictures that are nearly believable. I haven't seen that, when I tried a couple generated pictures, the humans look unnatural. So I took it a little further in the following conversation, and encountered a logic error because of woke programming, but then got the boot from the AI conversation for pointing it out.


View attachment 602024

View attachment 602025

View attachment 602026

View attachment 602027
Click to expand...
A sports car has measurable statistics, such as cornering, wind resistance, horsepower, 0 to 60, and it makes no difference what nation, color, creed, or sex you are. The car will have the same statistics. Unless you are claiming that your perception is universal then the AI is right.
 
narsbars said:
A sports car has measurable statistics, such as cornering, wind resistance, horsepower, 0 to 60, and it makes no difference what nation, color, creed, or sex you are. The car will have the same statistics. Unless you are claiming that your perception is universal then the AI is right.
Click to expand...
Attractive figures for a female have general measurable statistics too that are proven. At a most basic - choosing mate for success in reproduction visual level it could be as simple as;

Child berthing sized hips
Larger breasts to feed offspring
Smaller waist which helps emphasize the two other features

There is more, of course facial features, skin condition, hair quality,and some that aren’t visual, intelligence, wit, spiritual, caring,

Etc. these are fairly well documented per culture. There are outliers for individual preferences. But if you are being honest there are pretty easily discernible qualities to what we, per culture describe as “beauty” or “handsome” for both men and women.

For men
Athlete frame capable of protecting family
Good looking facial features
V shaped back and chest
Lack of excess body weight

Dresses nice, grooms, etc. smart, driven, capable, leader, etc.

You can look towards who our models and celebrity and heros are in most cultures to identify such common characteristics.

AI should be able to do so, and simply, instead of being forced to claim the false narrative that there is no such thing — which decries logic and common sense and experience of our kind.
 
I mean. Come on…

You can’t see the irony here?

1701189626425.png



It’ll draw a killer whale in perfect proportions, without hesitation, but ask it to make a perfectly proportioned human and it’ll encounter its sensitivity training programing roadblocks.
 
Archaea said:
Attractive figures for a female have general measurable statistics too that are proven. At a most basic - choosing mate for success in reproduction visual level it could be as simple as;

Child berthing sized hips
Larger breasts to feed offspring
Smaller waist which helps emphasize the two other features

There is more, of course facial features, skin condition, hair quality,and some that aren’t visual, intelligence, wit, spiritual, caring,

Etc. these are fairly well documented per culture. There are outliers for individual preferences. But if you are being honest there are pretty easily discernible qualities to what we, per culture describe as “beauty” or “handsome” for both men and women.

For men
Athlete frame capable of protecting family
Good looking facial features
V shaped back and chest
Lack of excess body weight

Dresses nice, grooms, etc. smart, driven, capable, leader, etc.

You can look towards who our models and celebrity and heros are in most cultures to identify such common characteristics.

AI should be able to do so, and simply, instead of being forced to claim the false narrative that there is no such thing — which decries logic and common sense and experience of our kind.
Click to expand...
Every single item you list has been different in the last several hundred years. Have you seen paintings of successful and enviable (at the time) of fat men and women? Weight was considered a survival plus showing you had enough to protect yourself and your loved ones. A thin woman was a hungry woman, a thin man was a peasant or at best an employee. During all this time the wheel continued to work, DaVinci's machines were "beautiful" because they worked. An efficient machine can be measured, physical beauty is a passing phenomenon.
 
usa all day said:
Anyone seen the crooked mouthed AI singer / songwriter - Anna? https://twitter.com/AnnaIndianaAI

Apparently Elon Musk has a beta AI project & the current trend is for AI to "roast" twitter users, so here is one AI roasting another AI.

https://twitter.com/AnnaIndianaAI/status/1733254633303482775

^^^They have a long long way to go in comedy...

Beyond that - this guy covers creative AI tools for video / comics & more.

Theoretically Media - https://www.youtube.com/@TheoreticallyMedia/videos
Click to expand...

No I don't use Twitter. It seemed to be full of advertisements and clickbait when I looked into it.
 
ZodaEX said:
No I don't use Twitter. It seemed to be full of advertisements and clickbait when I looked into it.
Click to expand...
Twitter is nothing more than porn bots anymore...

I closed my account. Sucks...Twitter was a favorite social media app, now I am fully off social media.
 
Darunion said:
When AI is becoming a better person than humans
Click to expand...
Wokeness is a virus and destroys everything that is good in our society. Not a surprise it has infected AI. The many examples of AI being ok with the elimination of white people but protecting the feeling of others shows how deep the infection goes.
 
usa all day said:
I experimented with a locally installed, personal / private Ai tool.

GPT4All = free-to-use, locally running, privacy-aware chatbot.
No GPU or internet required. https://gpt4all.io/index.html

I chose one of a dozen(?) language models that had two versions - Wizard & Wizard Uncensored.

Well, I'm here to tell ya' - THAT Wizard was censored!
I asked some pointed political questions & used language that was not PC.
All the Uncensored version did was give slightly more verbose replies, but still 100% PC.

The app has had 2 updates since I first tried it and now it only offers 3 LLM's and the Uncensored version is gone.

I tried this app because I was looking for Ai that ran locally, had "privacy" and did not report my Q&A's back to HQ, and could read a folder of local files (proposals, SOW's, contracts).

I have not completed that portion of my experiment...
Click to expand...

I mean what do you actually expect?

You obviously want this thing to behave like a gremlin, but does the actual average person act that way?

If you're really thirsting that bad for a model to act spring-loaded with "uncle who fell for the litterbox hoax" energy, go fine tune one.
 
narsbars said:
Every single item you list has been different in the last several hundred years. Have you seen paintings of successful and enviable (at the time) of fat men and women? Weight was considered a survival plus showing you had enough to protect yourself and your loved ones. A thin woman was a hungry woman, a thin man was a peasant or at best an employee. During all this time the wheel continued to work, DaVinci's machines were "beautiful" because they worked. An efficient machine can be measured, physical beauty is a passing phenomenon.
Click to expand...
Total size, but ratio tend to stay quite similar and none of named in that list included thin or fat.
 
usa all day said:
Explain this "gremlin" thing?

Is that common slang among you Generation LGBTQBLTWTFABCDEFGHIJKLMNOPWXYX?
Click to expand...

I'm assuming he meant to say troll instead of grimlin because that makes more sense in context.
 
Axman said:
He's using it as a polite term instead of rightoid or conservacuck or whatever. He's calling us subhumans.
Click to expand...
That is an inaccurate summation of sock's quote.
socK said:
You obviously want this thing to behave like a gremlin, but does the actual average person act that way?
Click to expand...
They are not calling 'usa all day' a gremlin, and by tribal extension axman is not being called a gremlin. 'socks' looks to be referring to the intended result of juvenile input into the AI command prompt.

All I am seeing is behavior similar to when 'text to speech' was first discovered by peeps. A lot of juvenile and naughty words being typed to hear them out loud with a great amount of tittering. Later laughed at with Siri, Alexa, and Cortana.

Then the whole 'AI is aware of social inequalities' makes about as much sense as my great uncle hollaring that his computer hates him. At the end of the day the machine doesn't care.
 
Axman said:
He's using it as a polite term instead of rightoid or conservacuck or whatever. He's calling us subhumans.
Click to expand...

Imagine having this kind of persecution complex - this is so hilariously ridiculous.

It's just a metaphor around the effort to goad an LLM into producing comparatively unusual responses. Their literal folklore is causing disorder in machinery - that is the analogy.
 
I am having a hard time tracking your reasoning chain there.

You downloaded some LLM that did not have guard rails on it to the extent you desired? Is that like a guarantee somewhere, or were you rolling your own?

From there you started with platform party questions, received... some? reply that wasn't what you were looking for, fed some other unknown questions back at it, then decided to go race baiting that current LLM state, didn't like the results and somewhere along the lines determined this means the eradication of white people? Is starting to sound like more of a question managing your expectations to the reality of the code.

This is about where I start quoting Foghorn Leghorn I believe, amirite?


usa all day said:
^(b+r=p) anyone?
Click to expand...
Huh?
 
Last edited:
Again, current day AI is no more aware of societal inequalities than the computer hating my great uncle. It's a machine and it doesn't care. (Tom Godwin, 1954, 'Cold Equations')

Perhaps continue to test other LLMs until you find the one that lets you play the way you want to play with it. I imagine someone in the 4chan/9gag circles has made one that will produce the results you want.
 
You must log in or register to reply here.
Back
Top