RUSSIA

DE4173 · August 17

Lerxst1992 said:

DE4173 said:

Not to derail off the topic of Russia further, but I don't know if doing a Google search and posting links or copying and pasting the findings are any different. Makes for a good discussion. Maybe?

I posted in some thread that if people search for stories for the view they want to portray, it may not tell the whole story on the subject(s).

Tim Simmons said:

It boils down to having someone else write your post.

Think of it like a term paper. Links/tweets/etc are like citations. AI is like buying your paper or paying someone to write it for you.

Is there discussion here DE? What the far left posts here are anecdotal news stories from one side. And emotional angry sarcastic posts.

no one here cares about solving problems, only here as a safe space to trumpet one side of the truth.

when I post ai, the purpose is to provide balance, information and data.

Majority is a blue MAGA circle jerk.

DE4173 · August 17

The citation:

https://www.dictionary.com/browse/circle-jerk

A circle jerk is when a group of men sit or stand in a circle and masturbate. Figuratively, a circle jerk is a group of people who are "getting themselves off" in the echo chamber of their own opinions or activities.

Halifax2TheMax · August 17

“Some scholars.” Like “some trolls.”

Tim Simmons · August 17

Lerxst1992 said:

DE4173 said:

Not to derail off the topic of Russia further, but I don't know if doing a Google search and posting links or copying and pasting the findings are any different. Makes for a good discussion. Maybe?

I posted in some thread that if people search for stories for the view they want to portray, it may not tell the whole story on the subject(s).

Tim Simmons said:

It boils down to having someone else write your post.

Think of it like a term paper. Links/tweets/etc are like citations. AI is like buying your paper or paying someone to write it for you.

when I post ai, the purpose is to provide balance, information and data.

But is it accurate?

Tim Simmons · August 17

Lerxst1992 said:

Halifax2TheMax said:

Tim Simmons said:

It boils down to having someone else write your post.

Think of it like a term paper. Links/tweets/etc are like citations. AI is like buying your paper or paying someone to write it for you.

Without the citations and foot notes or a bibliography.

”some scholars.”

SO WHAT, if that’s what AI quoted...

...the perfect example of what is wrong in this era, and why internet communications fail our society

you posted this back to back without any self awareness?!

FOH.

Halifax2TheMax · August 17

Tim Simmons said:

Lerxst1992 said:

DE4173 said:

Not to derail off the topic of Russia further, but I don't know if doing a Google search and posting links or copying and pasting the findings are any different. Makes for a good discussion. Maybe?

I posted in some thread that if people search for stories for the view they want to portray, it may not tell the whole story on the subject(s).

Tim Simmons said:

It boils down to having someone else write your post.

Think of it like a term paper. Links/tweets/etc are like citations. AI is like buying your paper or paying someone to write it for you.

when I post ai, the purpose is to provide balance, information and data.

But is it accurate?

AI music is better than pearl jam. AI recipes are better than your grandmother’s. Conversations with AI are better than with your friends and family as is advice on how to fix things. And AI is better than going to the doctor because you don’t even need to leave the house. Speaking of which, AI also offers outstanding financial advice so no need for a financial advisor and it can do my taxes. In fact, AI is some best friend and it’ll even provide spiritual advice too.

Tim Simmons · August 17

I think its becoming more and more clear this forum may not be for me.

benjs · August 17

Tim Simmons said:

I think its becoming more and more clear this forum may not be for me.

Amen.

gimmesometruth27 · August 17

benjs said:

Tim Simmons said:

I think its becoming more and more clear this forum may not be for me.

Amen.

just use the ignore feature.

benjs · August 17

gimmesometruth27 said:

benjs said:

Tim Simmons said:

I think its becoming more and more clear this forum may not be for me.

Amen.

just use the ignore feature.

I chose a different route. My subscription expires in November, and I turned off the auto-renew.

static111 · August 17

AI is not objective

Indifference · August 17

Indifference · August 17

j

HughFreakingDillon · 1:35AM

it won't matter that it's all horseshit. you'll believe it the instant you finish reading the title of the article/tweet/chyron.

Lerxst1992 · 3:11AM

Tim Simmons said:

Lerxst1992 said:

Halifax2TheMax said:

Tim Simmons said:

It boils down to having someone else write your post.

Think of it like a term paper. Links/tweets/etc are like citations. AI is like buying your paper or paying someone to write it for you.

Without the citations and foot notes or a bibliography.

”some scholars.”

SO WHAT, if that’s what AI quoted...

...the perfect example of what is wrong in this era, and why internet communications fail our society

you posted this back to back without any self awareness?!

FOH.

I’d love to work front of house. Thanks. Quoting “some scholars” is a lot more than what that commenter does

that dude opinion shops journalists to find someone that agrees with them. If you believe that has any “thought” or can compete with a language learning model that can comprehend massive and diverse data sets, so be it.

Lerxst1992 · 3:26AM

static111 said:

AI is not objective

Correct. they are influenced by data, not emotion. And there is nothing wrong examining summaries and extrapolations of vast amounts of data vs the old way of doing it…which was liberals looking to NYT and conservatives using WSJ…that is nothing more than confirmation bias. And that’s all that used to exist on forums like this one.

just look at the reactions recently, a few using this as an excuse to run away. From what? Reminds me how Schools got freaked out by this technology because they haven’t been challenged like this in the past. There is a way to use this technology to learn and to figure out whether their students are actually learning. But it takes a new approach, and ironically ai will be able to help educators figure out how.

….

Generally, no. Large language models (LLMs) are not objective. Their responses are influenced by the data they were trained on, which contains the biases, perspectives, and ideologies present in the vast amount of text and code from the internet and other sources.

What Makes LLMs Subjective?

LLMs don't have personal beliefs or consciousness, but they are built on a foundation of human-generated data, making them susceptible to replicating and amplifying existing biases. Think of an LLM as a very advanced pattern-matching system. It learns the statistical relationships between words and phrases from its training data. If the training data contains a disproportionate amount of information about certain groups or if it presents a specific point of view as the norm, the model will reflect that in its output.

* Training Data Bias: The most significant factor is the training data. A model trained on a dataset where certain perspectives are overrepresented will naturally favor those viewpoints. For example, a model might generate stereotypical content about different demographics if its training data contains such stereotypes.

* Reinforcement Learning from Human Feedback (RLHF): This process, used to fine-tune models, also introduces subjectivity. Human annotators rank and rate the model's responses, and the model learns to align with their preferences. Since these human raters have their own biases and cultural norms, their feedback can subtly guide the model toward certain subjective outcomes.

* Inherent Design: The very nature of language is subjective. Words and phrases can have different meanings and connotations depending on context, culture, and individual interpretation. Because LLMs are designed to generate human-like text, they inherently operate in a domain that is far from objective. This is similar to how abstract art, while following certain rules of composition, is inherently open to subjective interpretation.

What About "Factual" Information?

Even when an LLM provides what seems like a simple, factual answer, it's not truly objective. The model's response is a synthesis of information it has "seen," and it may present one version of events or a specific scientific theory without acknowledging alternative or less-common perspectives. It might also struggle with nuance, misinterpreting facts or presenting information out of context.

For a response to be truly objective, it would need to be entirely free of human perspective, interpretation, and bias. Given that LLMs are a product of human language and design, this level of objectivity is impossible for them to achieve.

Halifax2TheMax · 4:20AM

josevolution · 11:45AM

So today the president of this country will be bidding on behalf of Putin where Ukraine gets zero nada zilch but Russia gets to keep Crimea and Ukraine never gets into NATO! Now that’s a deal maker 🤮

Gern Blansten · 1:05PM

trump made so many promises related to Russia/Ukraine

1. He would end the war before taking office
2. He would end the war on day one
3. If Putin didn't agree to a cease fire he would "come down hard" on him

All complete bullshit. Not to mention giving this fucker the red carpet treatment on US soil.

You can tell how this negotiation went...Putin gives up nothing then trump says "will you at least say that you wouldn't have invaded Ukraine if I were potus?" and he gave him that. So pathetic.

Merkin Baller · 1:09PM

JointheDots on X GovPressOffice This image of the US soldiers on their knees rolling out the red carpet for Vladimir Putin will go down as one of the most shocking and shameful

American troops rolling a red carpet for Putin rMilitary

RUSSIA

Comments

Categories