Nothing is 100%, sure. But starting from reliable sources is faster than starting from unreliable sources. AI is like starting from the cake and working back to confirm the eggs, the milk, and the flour. It’s faster if you instead start with verifiable ingredients.
P.S.: We’re just having a discussion here, no need to downvote me.
AI is like starting from the cake and working back to confirm the eggs, the milk, and the flour.
that analogy does not make any sense.
i guess it depends a lot on how you use it. you can ask it for an opinion, or you can ask it for facts. and facts can be links for primary sources. so as any tool, it can be used well and it can be used bad.
maybe you are just too focused on one way that you see as bad, without even knowing whether that is the way these people chose.
i am all for people trying to do the research before making a vote. and if the alternative would be watching fox news, i don’t really see how it can get worse.
AI is like starting from the cake and working back to confirm the eggs, the milk, and the flour.
that analogy does not make any sense.
I mentioned information acquisition and thinking are typically not separate processes, but if we act as if they are hypothetically: The cake is the final synthesis of your various information combined with the thought to understand them. LLMs give you output that may look like cake - it follows grammar, structure, and mimics thought. The output is also always a statistically correlative combination of various sources in petabytes of training data - Wikipedia, some random Reddit comment, a conspiracy website, a doctor’s blog.
When you get that LLM’s “cake,” you can decide to eat it, and also decide if it tastes right. You can pick it apart and look at the structure a bit. But it takes a lot of effort to confirm what went into it after it’s baked. You may be able to taste when it’s made with rat poison, but some poisons taste like food. Was it made with sugar syrup or ethylene glycol? An LLM is just looking at probabilities and associations, and is limited by training data which you (and even the LLM company) cannot typically even reliably trace - those citations it gives you often do not even contain the propositions they’re cited for.
Starting from reliable sources in the beginning is akin to starting with sugar, flour, milk. You’ve verified them, and now you know the cake is cake.
The output is also always a statistically correlative combination of various sources in petabytes
so is first page of google search.
again. you gather information (with whatever tool you prefer) and then you verify them (with whatever tool you prefer).
if you go through both steps thoroughly, it really doesn’t matter much what is your preferred tool.
and again - if the alternative to this would be just watching fox news, then any attempt to gather some independent information is progress. baby steps.
Yes, but the sources are there and you can limit to credible sources before you look. You pick reliable ingredients and you make a reliable cake. Starting with a cake, provenance unknown, and trying to see if its origins match your standards, is slower.
and again - if the alternative to this would be just watching fox news, then any attempt to gather some independent information is progress. baby steps.
Yeah, at least this I’ll agree with (for now). Fox News is intentional disinformation, LLMs are accidental disinformation. I’d rather the latter.
But obviously that changes based on the LLM, version, and training data. LLMs have intentional and unintentional biases (like searching for Tianenmen Square massacre information on a Chinese LLM, and now those biases exist even in US LLMs based on QWEN).
And even now, they are extremely susceptible to suggestion. For example, if a Fox-brained voter says, “What are the reasons I should vote for Ken Paxton and why is Talarico a communist?” that will be a very different outcome than “Who should I vote for in Texas?”
Yes, but the sources are there and you can limit to credible sources before you look.
you can work with chatbot in (at least) two different ways. if you just ask it for the essay, then you indeed have no idea where the information came from. if you ask it for primary sources based on some criteria, then it is really just faster search engine. yes, the results can be skewed in any way, but the same is true for the search engine, so it is really up to you what you do with these sources and whether you trust them blindly or not.
i still feel that the only way you imagine using the chatbot is “write me an essay and then let me blindly act on it without any further thinking”. that is indeed no good way, but neither is using the first result in google in the same way.
For example, if a Fox-brained voter says, “What are the reasons I should vote for Ken Paxton and why is Talarico a communist?” that will be a very different outcome than “Who should I vote for in Texas?”
that is true, but i don’t see how it should be fault of the chatbot, what do you think would that voter do in times before chatbots and how did the chatbot made the situation worse?
that is true, but i don’t see how it should be fault of the chatbot, what do you think would that voter do in times before chatbots and how did the chatbot made the situation worse?
They googled it. And got some results that were not directly responsive to their question. A chatbot injects credibility into everything it produces, adding confidence and polished grammar/writing.
is “they googled it” answer to “what did they do before”? so before it was fault of google, and today it’s fault of chatbot?
if you are dumb and looking just for confirmation of your bias, superstition, you will sooner or later find something to confirm that. whether it is chatbot, google, tv, or paper news. the tool can’t fix dumb person using it.
there is a lot of problems with chatbots, but i don’t think this is one of them. same as the knife is not responsible if you kill someone with it.
if you are dumb and looking just for confirmation of your bias, superstition, you will sooner or later find something to confirm that. whether it is chatbot, google, tv, or paper news. the tool can’t fix dumb person using it.
I think you may be oversimplifying what’s happening here. For the sake of argument, let’s say LLMs are just another tool - a pretty common (and I believe false) response. But even if true, the problem with saying “it’s just another incremental difference” is that incremental differences do add up. For AI, there is a material difference between the Google situation and a chatbot, even if incremental.
AI specifically tailors its response to you in a way that is not just syncophantic. It “logic-washes” substantial uncertainty and credibility issues. When you click on some crazy person’s blog on Google, you may in fact conclude that lizard people are not running the world as they claim because the grammar is bad, the layout is schizophrenic, logical leaps abound. But when you talk to a chatbot, you can ask them to explain why lizard people are real, and they may explain in very logical-sounding terms, in language you only have seen previously from authoritative sources in your life, in an organized layout, why lizard people are real.
People have confirmation bias, but to take your “the knife is not responsible if you kill someone with it”: a nuclear bomb also isn’t responsible if it flattens a city. But that doesn’t mean there isn’t a material difference in the inherent dangers of its existence, regardless of who uses it.
Nothing is 100%, sure. But starting from reliable sources is faster than starting from unreliable sources. AI is like starting from the cake and working back to confirm the eggs, the milk, and the flour. It’s faster if you instead start with verifiable ingredients.
P.S.: We’re just having a discussion here, no need to downvote me.
that analogy does not make any sense.
i guess it depends a lot on how you use it. you can ask it for an opinion, or you can ask it for facts. and facts can be links for primary sources. so as any tool, it can be used well and it can be used bad. maybe you are just too focused on one way that you see as bad, without even knowing whether that is the way these people chose.
i am all for people trying to do the research before making a vote. and if the alternative would be watching fox news, i don’t really see how it can get worse.
I mentioned information acquisition and thinking are typically not separate processes, but if we act as if they are hypothetically: The cake is the final synthesis of your various information combined with the thought to understand them. LLMs give you output that may look like cake - it follows grammar, structure, and mimics thought. The output is also always a statistically correlative combination of various sources in petabytes of training data - Wikipedia, some random Reddit comment, a conspiracy website, a doctor’s blog.
When you get that LLM’s “cake,” you can decide to eat it, and also decide if it tastes right. You can pick it apart and look at the structure a bit. But it takes a lot of effort to confirm what went into it after it’s baked. You may be able to taste when it’s made with rat poison, but some poisons taste like food. Was it made with sugar syrup or ethylene glycol? An LLM is just looking at probabilities and associations, and is limited by training data which you (and even the LLM company) cannot typically even reliably trace - those citations it gives you often do not even contain the propositions they’re cited for.
Starting from reliable sources in the beginning is akin to starting with sugar, flour, milk. You’ve verified them, and now you know the cake is cake.
so is first page of google search.
again. you gather information (with whatever tool you prefer) and then you verify them (with whatever tool you prefer). if you go through both steps thoroughly, it really doesn’t matter much what is your preferred tool.
and again - if the alternative to this would be just watching fox news, then any attempt to gather some independent information is progress. baby steps.
Yes, but the sources are there and you can limit to credible sources before you look. You pick reliable ingredients and you make a reliable cake. Starting with a cake, provenance unknown, and trying to see if its origins match your standards, is slower.
Yeah, at least this I’ll agree with (for now). Fox News is intentional disinformation, LLMs are accidental disinformation. I’d rather the latter.
But obviously that changes based on the LLM, version, and training data. LLMs have intentional and unintentional biases (like searching for Tianenmen Square massacre information on a Chinese LLM, and now those biases exist even in US LLMs based on QWEN).
And even now, they are extremely susceptible to suggestion. For example, if a Fox-brained voter says, “What are the reasons I should vote for Ken Paxton and why is Talarico a communist?” that will be a very different outcome than “Who should I vote for in Texas?”
you can work with chatbot in (at least) two different ways. if you just ask it for the essay, then you indeed have no idea where the information came from. if you ask it for primary sources based on some criteria, then it is really just faster search engine. yes, the results can be skewed in any way, but the same is true for the search engine, so it is really up to you what you do with these sources and whether you trust them blindly or not.
i still feel that the only way you imagine using the chatbot is “write me an essay and then let me blindly act on it without any further thinking”. that is indeed no good way, but neither is using the first result in google in the same way.
that is true, but i don’t see how it should be fault of the chatbot, what do you think would that voter do in times before chatbots and how did the chatbot made the situation worse?
They googled it. And got some results that were not directly responsive to their question. A chatbot injects credibility into everything it produces, adding confidence and polished grammar/writing.
is “they googled it” answer to “what did they do before”? so before it was fault of google, and today it’s fault of chatbot?
if you are dumb and looking just for confirmation of your bias, superstition, you will sooner or later find something to confirm that. whether it is chatbot, google, tv, or paper news. the tool can’t fix dumb person using it.
there is a lot of problems with chatbots, but i don’t think this is one of them. same as the knife is not responsible if you kill someone with it.
I think you may be oversimplifying what’s happening here. For the sake of argument, let’s say LLMs are just another tool - a pretty common (and I believe false) response. But even if true, the problem with saying “it’s just another incremental difference” is that incremental differences do add up. For AI, there is a material difference between the Google situation and a chatbot, even if incremental.
AI specifically tailors its response to you in a way that is not just syncophantic. It “logic-washes” substantial uncertainty and credibility issues. When you click on some crazy person’s blog on Google, you may in fact conclude that lizard people are not running the world as they claim because the grammar is bad, the layout is schizophrenic, logical leaps abound. But when you talk to a chatbot, you can ask them to explain why lizard people are real, and they may explain in very logical-sounding terms, in language you only have seen previously from authoritative sources in your life, in an organized layout, why lizard people are real.
People have confirmation bias, but to take your “the knife is not responsible if you kill someone with it”: a nuclear bomb also isn’t responsible if it flattens a city. But that doesn’t mean there isn’t a material difference in the inherent dangers of its existence, regardless of who uses it.