@wildgrass saidThe attacks are never necessary; they only cloud the discussion into personalities, removing the original intent. Avoiding the topic altogether to express contempt towards someone else, the topic is lost from that point on. It would be better to agree to disagree; at least, there, you are treating the other person with dignity instead of contempt.
A simple moment of self-reflection and self-correction fixes this, so no we cannot all fall into an echo chamber. Just ask someone you disagree with to explain themselves, listen, ask for evidence, and present your own counter-argument that also includes evidence. Personal attacks are only necessary if the other person is deliberately distorting reality by presenting false ...[text shortened]... ive and negative, make the case with data. The best evidence makes for the most convincing argument.
@KellyJay saidSure but I think an intention to resolve the argument should always be the goal. Otherwise why even come to the debate?
The attacks are never necessary; they only cloud the discussion into personalities, removing the original intent. Avoiding the topic altogether to express contempt towards someone else, the topic is lost from that point on. It would be better to agree to disagree; at least, there, you are treating the other person with dignity instead of contempt.
You can actually prove if someone is arguing in bad faith. If someone follows predictable patterns which either ignore obvious evidence or provide deceptive evidence from their end then it becomes obvious that there is no good faith argument to be had with that person.
An example??
One time Mott the Hoople was convinced that "furries" were coming to school dressed as cats and teachers were enabling them to defecate in litter boxes that were placed in the classroom. When I asked how he knew this, he posted a link to an anonymous comment from an alleged grandmother of one of the kids who went to that school. So, on the one hand we have teachers and students and parents and superintendents and janitors, in addition to common sense, all pointing in one direction while a single anonymous grandma points the other way.
Agree to disagree isn't the end point here. Mott was just wrong.
@wildgrass saidI do not mind anyone sticking to their guns, explaining what they think and why. I do not call them names if, at any time during our discussion, they do not agree with me. You can assume a lot and be way off base. How many times have you seen or said, “You only say that because...” That is bad faith, assuming one can read the other person’s motivation and lump everything about that person’s point of view into whatever group they are being charged with belonging to. That is not discussing in good faith. Let Mott the Hoople deal with Mott the Hoople's posts. I will answer for what I say, not anyone else.
Sure but I think an intention to resolve the argument should always be the goal. Otherwise why even come to the debate?
You can actually prove if someone is arguing in bad faith. If someone follows predictable patterns which either ignore obvious evidence or provide deceptive evidence from their end then it becomes obvious that there is no good faith argument to be had w ...[text shortened]... us grandma points the other way.
Agree to disagree isn't the end point here. Mott was just wrong.
@KellyJay saidFine, whatever. The point of this thread was about the folly of trusting AI chatbots that are being co-opted and reprogrammed by corporations and third parties to make you think a certain way, and almost no one is checking on the source or the evidence. The ultimate effect is that it rots people's brains into trusting the computer to do your thinking for you.
I do not mind anyone sticking to their guns, explaining what they think and why. I do not call them names if, at any time during our discussion, they do not agree with me. You can assume a lot and be way off base. How many times have you seen or said, “You only say that because...” That is bad faith, assuming one can read the other person’s motivation and lump everything a ...[text shortened]... et Mott the Hoople deal with Mott the Hoople's posts. I will answer for what I say, not anyone else.
Don't let that happen to you. Weigh all the evidence. If someone ask you how you know something, and you have to grab your phone and ask a robot to give you the evidenced you need to make the claim that you had already decided on, then you've lost the argument.
@Mott-The-Hoople saidWhile most of your arguments are simply a series of stupid assumptions (and lies).
“Thinking people, if they can articulate why they believe what they do, can put together an argument; however, those driven by feelings, you immediately see their emotions, not their thoughts, on a topic.”
Exactly…that is why most libs on here devolve into name calling and perverted sexual terms, they can’t sustain a rational argument.
@wildgrass saidFine, whatever, I agreed from the start about why I thought that was a bad idea. Letting the computer think for you or staying with the herd mentality, I don’t see any difference if you are not considering all possibilities; instead of just those that prove your point, you lose the ability to be corrected.
Fine, whatever. The point of this thread was about the folly of trusting AI chatbots that are being co-opted and reprogrammed by corporations and third parties to make you think a certain way, and almost no one is checking on the source or the evidence. The ultimate effect is that it rots people's brains into trusting the computer to do your thinking for you.
Don't let t ...[text shortened]... evidenced you need to make the claim that you had already decided on, then you've lost the argument.
@Soothfast saidThe trouble AI has, much like us, is how to correct itself when it is wrong. If our formulas are designed to give us an answer based on our input, we will get an answer, even if the answer is based on incorrect information. Without a way to check, we will only be able to validate our bad answers with bad information that got us there in the first place.
The long-term problem is that AI is using Internet content to produce Internet content, and as the Internet becomes more and more made up of AI crap, AI will be eating its own crap to produce still crappier crap. It is not a sustainable model.
@KellyJay said
The trouble AI has, much like us, is how to correct itself when it is wrong. If our formulas are designed to give us an answer based on our input, we will get an answer, even if the answer is based on incorrect information. Without a way to check, we will only be able to validate our bad answers with bad information that got us there in the first place.
Without a way to check
There is a way to check. You just need to read the source material, check that it matches with what the AI spits out from a valid source (rather than a reddit post). But only 1 in 200 people do that. Also, you can ask AI to present a conflicting theory or any evidence that its conclusion is wrong. But again, very few people do that.
1 edit
@wildgrass saidThe only constraint truth has is reality itself. Truth only changes as reality does, and neither of those has anything to do with human opinion. AI gets its info from humans. Human opinions are based on our understanding, and that understanding is an ever-increasing pool. Still, it is a mixed bag: some are factual truths where we get it right, and others are opinions treated as if they were facts, which is unavoidable. Even a very small conceptual error propagates and does not remain small; it affects all downstream reasoning, making it impossible to get a correct understanding. Depending on what is under review, those errors could throw off everything in any number of areas; therefore, unless you know your source material is reality-based only, you stand just as likely to validate errors as not. Any error in opinion treated as if it were factual spoils the conclusions, so that they can be used to invalidate true statements or misidentify errors, leading us to believe in our errors as factual. This is on top of what you just said, which I completely agree with.Without a way to check
There is a way to check. You just need to read the source material, check that it matches with what the AI spits out from a valid source (rather than a reddit post). But only 1 in 200 people do that. Also, you can ask AI to present a conflicting theory or any evidence that its conclusion is wrong. But again, very few people do that.
@KellyJay saidOk, now you veer into philosophical arguments. Facts are the basis for drawing conclusions, but you make it seem like a conclusion is an opinion? These are different things not to be conflated.
The only constraint truth has is reality itself. Truth only changes as reality does, and neither of those has anything to do with human opinion. AI gets its info from humans. Human opinions are based on our understanding, and that understanding is an ever-increasing pool. Still, it is a mixed bag: some are factual truths where we get it right, and others are opinions tre ...[text shortened]... ieve in our errors as factual. This is on top of what you just said, which I completely agree with.
But let's say you wanted answers to a concrete question. You ask AI for the answer. Do you check the source of the computer's information? Because in many cases, the sources are being planted by people (or other computers) who are trying to influence the answers provided by Gemini, ChatGPT etc.
@wildgrass saidWhere do you think AI get it’s fact, from sources with nothing in common with humanity?
Ok, now you veer into philosophical arguments. Facts are the basis for drawing conclusions, but you make it seem like a conclusion is an opinion? These are different things not to be conflated.
But let's say you wanted answers to a concrete question. You ask AI for the answer. Do you check the source of the computer's information? Because in many cases, the sources are b ...[text shortened]... people (or other computers) who are trying to influence the answers provided by Gemini, ChatGPT etc.
@KellyJay saidAI doesn't get facts from anywhere.
Where do you think AI get it’s fact, from sources with nothing in common with humanity?
AI reads text and generates text. You need to verify that the sources are trustworthy, that the data was correctly interpreted, and that the conclusions match with reality.