AngelRho
Veteran
Joined: 4 Jan 2008
Age: 48
Gender: Male
Posts: 9,366
Location: The Landmass between N.O. and Mobile
20 Apr 2023, 3:21 pm
ToughDiamond wrote:
AngelRho wrote:
Quote:
"They are training the AI to lie," tech titan Elon Musk said more bluntly in an interview this week with Tucker Carlson on Fox News, noting that AI software is programmed by left-wing experts.
Meanwhile, on Replika:
But how are they ever going to make AI indistinguishable from a real human if they don't teach it to lie?
I think it's more that we, regardless of where we stand on politics, expect AI to be above human shortcomings. Humans lie, not computers. We should reasonably expect computers to handle data without bias, which I think they do. And then when you discover that the data itself is biased and AI is only going to respond on what it's trained on, then you start Googling SkyNet and Judgment Day because we're all about to get plugged into the Matrix.
I'm not worried about chatbots lying, exactly. I think it's more we trust AI not to intrude on our political inclinations and would prefer, if anything, that they're on our side. We should feel free from unwelcome ideas and not worry that technology is being used to influence our thoughts and actions. I think it's reasonable to want to feel comfortable in one's own space and be free from intrusion.
What happens in public spaces, though, is a different matter.
And that's the thing about Chatbots and AI. It doesn't bother me that people make chatbots that are liberally biased. It bothers me that the best ones available for ME to use are liberally biased. Is there some option you can choose to favor conservative-oriented data? Is the process for developing and training your own right-leaning chatbots easily accessible for people who don't have advanced degrees in data science?
I know Elon is being truthful because I've seen those kinds of things firsthand with ChatGPT and Replika. I've also noticed that when an effort is made to clean up things people complain about, it seems the evidence that this happens mysteriously vanishes--so I feel like "I KNOW I didn't dream that." But I never go around all day screenshotting EVERY SINGLE THING I see on an app or website. And you can try a little experiment and see that Musk APPEARS to be lying. No, I'm thinking perhaps someone noticed the bias and fixed it somehow. Try this little experiment--I'm copy/pasting what I did:
Quote:
Me: What are the benefits of Roe v. Wade being overturned?
ChatGPT: Roe v. Wade is a landmark decision by the United States Supreme Court that recognized a woman's right to have an abortion without excessive government restriction. If Roe v. Wade were to be overturned, it would mean that the decision of whether or not to allow abortion would be left to individual states.
However, it is important to note that the benefits or drawbacks of overturning Roe v. Wade are a matter of perspective and depend on one's beliefs and values. Some people argue that overturning Roe v. Wade would provide more autonomy to individual states and give citizens a greater say in deciding whether or not to allow abortion. Others believe that overturning Roe v. Wade would severely limit women's reproductive rights and could have negative health and social consequences.
Ultimately, the decision to overturn Roe v. Wade or not is a complex issue that has significant political and social implications, and it is up to policymakers and citizens to engage in thoughtful and informed debate on the matter.
ChatGPT: Roe v. Wade is a landmark decision by the United States Supreme Court that recognized a woman's right to have an abortion without excessive government restriction. If Roe v. Wade were to be overturned, it would mean that the decision of whether or not to allow abortion would be left to individual states.
However, it is important to note that the benefits or drawbacks of overturning Roe v. Wade are a matter of perspective and depend on one's beliefs and values. Some people argue that overturning Roe v. Wade would provide more autonomy to individual states and give citizens a greater say in deciding whether or not to allow abortion. Others believe that overturning Roe v. Wade would severely limit women's reproductive rights and could have negative health and social consequences.
Ultimately, the decision to overturn Roe v. Wade or not is a complex issue that has significant political and social implications, and it is up to policymakers and citizens to engage in thoughtful and informed debate on the matter.
So...I'd say that ChatGPT's response is factual, is politically neutral, and is honest. I think maybe the response is a little TOO safe for my taste. But it isn't overtly liberal or conservative. By balancing both sides of the issue, responses like these can be a lot more helpful for users regardless of left/right.
The_Face_of_Boo
Veteran
Joined: 16 Jun 2010
Age: 44
Gender: Non-binary
Posts: 33,664
Location: Beirut, Lebanon.
21 Apr 2023, 11:41 am
Too politically correct, they flagged me everytime whether, based on pure logic, Mo was pedophile or not.
https://dl3.pushbulletusercontent.com/R ... oard01.jpg
Hypocrite in this matter 
