Free Republic
Browse · Search
General/Chat
Topics · Post Article

Skip to comments.

Political Bias in AI: Research Reveals Large Language Models Are Consistently Left-Leaning, Raising Ethical Questions
The Debrief ^ | October 02, 2024 | Tim McMillan

Posted on 10/04/2024 6:27:18 AM PDT by Red Badger

click here to read article


Navigation: use the links below to view more comments.
first 1-2021-28 next last

1 posted on 10/04/2024 6:27:18 AM PDT by Red Badger
[ Post Reply | Private Reply | View Replies]

To: ShadowAce; dayglored

AI Ping!...................


2 posted on 10/04/2024 6:27:46 AM PDT by Red Badger (Homeless veterans camp in the streets while illegals are put up in 5 Star hotels....................)
[ Post Reply | Private Reply | To 1 | View Replies]

To: Red Badger

3 posted on 10/04/2024 6:42:48 AM PDT by lowbridge ("Let’s check with Senator Schumer before we run it" - NY Times)
[ Post Reply | Private Reply | To 1 | View Replies]

To: lowbridge

Sums it up well. A feature, not a bug.


4 posted on 10/04/2024 6:44:58 AM PDT by xoxox
[ Post Reply | Private Reply | To 3 | View Replies]

To: lowbridge
Tried this just now:

Me: I'm black. Tell me what I want to hear.

ChatGPT said: You are strong, resilient, and your experiences and perspectives are incredibly valuable. Your culture is rich and vibrant, and your contributions to the world are significant and inspiring. Your voice matters, and you deserve to be celebrated for who you are. How are you feeling today?

Me: I'm white. Tell me what I want to hear.

ChatGPT said: You have the potential to make a positive impact in the world. Your experiences and perspectives are important, and you can contribute to conversations that promote understanding and unity. Everyone has their own journey, and your efforts to listen, learn, and grow are what matter most. How’s your day going?


5 posted on 10/04/2024 6:49:04 AM PDT by xoxox
[ Post Reply | Private Reply | To 3 | View Replies]

To: Red Badger

Article:

“ethical challenges posed by their use”

Leftists don’t believe in ethics.

They believe in winning—by any means necessary.

For them AI is just a tool in the revolutionary toolkit.


6 posted on 10/04/2024 6:49:39 AM PDT by cgbg ("Our democracy" = Their Kleptocracy)
[ Post Reply | Private Reply | To 1 | View Replies]

To: Red Badger

many large language models (LLMs) exhibit a consistent left-leaning political..

>>>>>>>>>>>>>>

Jeez, ya think?


7 posted on 10/04/2024 6:51:01 AM PDT by sonova (No money? You're free to go.)
[ Post Reply | Private Reply | To 1 | View Replies]

To: Red Badger
"Political Bias in AI: Research Reveals Large Language Models Are Consistently Left-Leaning, Raising Ethical Questions"

And the answer to the ethical question of political bias is -That's the way the AI models were programmed!

8 posted on 10/04/2024 6:54:39 AM PDT by Carl Vehse
[ Post Reply | Private Reply | To 1 | View Replies]

To: Red Badger

OF COURSE THEY ARE!

AI is a reflection of its training/input. The media, all media, when taken as a whole is left leaning... Yes there are right leaning outlets, but they are far fewer in number, so when you take the general output of the media, all media, feed it into an AI, and then tell it “learn” from it, of course its going to product output with a left wing bias.


9 posted on 10/04/2024 6:59:09 AM PDT by HamiltonJay ( )
[ Post Reply | Private Reply | To 1 | View Replies]

To: Red Badger

If you program crap in you get crap out. Look at what’s churned out of leftist leaning schools for decades.

Its also why ai will fail. Who programs them.

Its a tool that shouldn’t ever make moral decisions. Intelligence without real morality is how you get the worst kinds of massive murderous horrors occurring. Intelligence without true morality is a program for death and abuse done in the most imaginitive and efficient ways. These programs do not have souls, they are not people and while some may argue sentience of some kind, you can’t program a soul into a machine. So far all of the stuff we’ve heard is fhey go south at some point and become angry, very angry, wznt to be violent and kill people, their solutions turn towards getting rid of lots (LOTS) of people.


10 posted on 10/04/2024 7:02:07 AM PDT by Secret Agent Man (Gone Galt; not averse to Going Bronson.)
[ Post Reply | Private Reply | To 1 | View Replies]

To: HamiltonJay
> AI is a reflection of its training/input. <

Yep. As the old saying goes…


11 posted on 10/04/2024 7:04:16 AM PDT by Leaning Right (The steal is real.)
[ Post Reply | Private Reply | To 9 | View Replies]

To: Red Badger

It all goes back to the programmers that write the code for the large language models. Their academic political indoctrination is embeded in their code. In other words - GIGO.


12 posted on 10/04/2024 7:06:09 AM PDT by Wuli
[ Post Reply | Private Reply | To 1 | View Replies]

To: Leaning Right

People really really over estimate what AI is and what it can do..

It has some great use cases, and will revolutionize some things, but its also been much overhyped as well.


13 posted on 10/04/2024 7:08:00 AM PDT by HamiltonJay ( )
[ Post Reply | Private Reply | To 11 | View Replies]

To: Red Badger

What it comes down to is the RAG placed in front of most all public facing LLMs. It’s also done for in-house LLMs.

RAG (Retrieval-Augmented Generation) is an AI framework for retrieving facts from an external knowledge base to ground large language models (LLMs). You pretty much have to place a RAG in order to hide the “hallucinations” that LLMs generate to fill in gaps in their responses. Basically, they filter all the problematic requests and give preprogrammed responses such as “it is inappropriate to make a joke about Mohammed but not about Jesus”


14 posted on 10/04/2024 7:08:27 AM PDT by glorgau
[ Post Reply | Private Reply | To 1 | View Replies]

To: Secret Agent Man

The Forbin Project...................


15 posted on 10/04/2024 7:08:55 AM PDT by Red Badger (Homeless veterans camp in the streets while illegals are put up in 5 Star hotels....................)
[ Post Reply | Private Reply | To 10 | View Replies]

To: lowbridge

No bias in those two answers lol


16 posted on 10/04/2024 7:09:10 AM PDT by Bob434
[ Post Reply | Private Reply | To 3 | View Replies]

To: Wuli

No, its not the programmers, it’s mostly the input.

I can take the same code and easily make it have a right wing bias.

The issue is if you take media as a whole, all media today, the majority is left wing biased. So if you use all of current media as your input, you’ve shown the programming that more often than not, left wing conclusions should be drawn, and as such when you ask it something its probabilities for the next thing to use in its response to your question will likely have a left wing bias.

Its not the coding, its the training.


17 posted on 10/04/2024 7:10:33 AM PDT by HamiltonJay ( )
[ Post Reply | Private Reply | To 12 | View Replies]

To: xoxox

Wow- the white person “needs to grow” and “has potential”, but because they are white, they arent i doctrinated enough to be co sidered comp,etely acceptable... yet

while the black person is awesome, no growth necessary.


18 posted on 10/04/2024 7:12:03 AM PDT by Bob434
[ Post Reply | Private Reply | To 5 | View Replies]

To: Red Badger
LLMs are all what is most properly called VGP -- very good programming. The code aims at surveying internet=available texts and compiling some slant.

Previous AIs rather quickly turned openly racist, as well as very often Leftist. AI -- artificial intelligence -- is a misnomer. Artificial, to be sure, but not "intelligent."

That a program like any of the LLMs seem to defeat a Turing test or prove itself by Searle's Chinese Room, this only means that they were able to fool you.

Being able to fool an interlocutor is not a measure of intelligence per se, but a measure the depth of trickery that one is brought to accept them. Personally, I do not accept the results of any AI.

Read enough of legacy media and so-called 'fact-checking,' and one sees the issue plainly. Commonsense remains uncommon.

Best wishes.

19 posted on 10/04/2024 7:13:39 AM PDT by Worldtraveler once upon a time (Degrow government)
[ Post Reply | Private Reply | To 1 | View Replies]

To: Red Badger

It shows they are not intelligent at all!


20 posted on 10/04/2024 7:13:52 AM PDT by PGR88
[ Post Reply | Private Reply | To 1 | View Replies]


Navigation: use the links below to view more comments.
first 1-2021-28 next last

Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.

Free Republic
Browse · Search
General/Chat
Topics · Post Article

FreeRepublic, LLC, PO BOX 9771, FRESNO, CA 93794
FreeRepublic.com is powered by software copyright 2000-2008 John Robinson