As is the sample in this article. You will surely find as many physicists saying earth is flat, mathematicians who hold that Cantor was wrong, and medical doctors that tell you vaccination against measles is overall worse than not.
Today I heard two negative cases from a friend interacting her clients. The first client replied to her emails by sending ChatGPT logs, apparently unable to communicate on his own. The second believed a ChatGPT hallucination that he was entitled to a special Amazon business account. She had to explain to him that ChatGPT will tell him convincing bullshit, which was news to him. One can only wonder what terrible choices and wrong beliefs he had made up until the moment of enlightenment.
As a consultant and subject matter expert, pretty every week, I have to make billable detours from productive work to explain to a client how they've been led astray by some ChatGPT response that they put errant trust in. It's been critically wrong pretty much any time a client's presented me with a response from it. I get paid to help when this haple s and to fix things that go wrong, so I guess technically that's not a negative from my perspective, but it is kind of frustrating and does seem pretty wasteful.
Meanwhile, as a friend, I've consistently had to coach people through doubting what they've received as medical guidance from ChatGPT and other chatbot or search LLM's, variously pointing out how: (a) the response doesn't actually agree with the cited sources nor correct information, (b) the cited are poor authorities (blogspam) and are not correct themselves, or (c) both.
Thankfully, the consequences in both these kinds of scenarios have been innocuous to date, so ostensibly "not negative", because most ChatGPT/chatbot inquiries are really for moot trivia rather than anything of true consequence, but it's repeatedly shown itself to be a pretty risky and unreliable tool, as things go, so I accept that it's only a matter of time before the true "negative experience" comes and nudge people away from it when I can.
The worse thing is the companies behind those tools know it, they know it’s not reliable and it looks like they want to build a world where a majority of people won’t check sources anymore and blindly trust the LLM as authority. I fear that, by a tour de force of some kind, they’ll shield themselves from giving any source is the future in the name of « intelligence ». What will happen in the next generations where young that have grown up only with LLM giving them answers they saw as truth and older generations died? What a different world it will be.
Just like social media bigshots don't allow their kids to use them, tech people using the oldest Thinkpad they can find running debian or a drug dealer doesn't consume his own product.
I have, but only because I’ve made it a point to learn about American history through contemporary sources.
For that matter, I live near the largest remaining KKK organization in the country; as you can imagine, I was exposed to a ton of racist language as a child. That’s a big part of why I spent so much time in early adulthood do historical research. I could probably list more than a dozen racial epithets for Black people that I’ve heard first-hand, but that one in particular isn’t among them.
This article isn't much better than the genre of "Some rando on Twitter said" style articles. They're interviewing AI raters, of which there are millions in the world. It's not difficult to find a half dozen in any group that large to espouse any position. Some kind of polling would at least provide some statistical rigour but just "we managed to find a few people who espouse this position" is there purely to reinforce people's pre-existing biases.
I found the article frustrating for the lack of examples of bad things AI does that they are avoiding. They say one avoids talking about palestinians but a lot of humans including myself do that as it's so easy to offend people. Also that they may not have censored all bad words and make mistakes occasionally but that applies to pretty much all media.
Not saying that AI doesn't have issues but the article makes a very vague case.
I noticed one of my engineers blindly trusts whatever AI says. When his teammates challenged him and showed actual data, he just went, “Oh yeah, the AI was wrong.”
For what it's worth, I have found AI useful. I have not had any negative experiences with Chat GPT.
Not to be sarcastic but the sample of your study is quite light. (I too I found it useful for my particular use case but that doesn’t say much either)
To be honest, the sample of the Guardian's study is also quite light (a dozen people) and with much more selection bias (people who work with AI).
As is the sample in this article. You will surely find as many physicists saying earth is flat, mathematicians who hold that Cantor was wrong, and medical doctors that tell you vaccination against measles is overall worse than not.
So? People sharing their experiences is good. Not everything is a scientific study.
True and opinions are often discarded as irrelevant in internet discussions in favor of large-scale studies.
Today I heard two negative cases from a friend interacting her clients. The first client replied to her emails by sending ChatGPT logs, apparently unable to communicate on his own. The second believed a ChatGPT hallucination that he was entitled to a special Amazon business account. She had to explain to him that ChatGPT will tell him convincing bullshit, which was news to him. One can only wonder what terrible choices and wrong beliefs he had made up until the moment of enlightenment.
Cool.
As a consultant and subject matter expert, pretty every week, I have to make billable detours from productive work to explain to a client how they've been led astray by some ChatGPT response that they put errant trust in. It's been critically wrong pretty much any time a client's presented me with a response from it. I get paid to help when this haple s and to fix things that go wrong, so I guess technically that's not a negative from my perspective, but it is kind of frustrating and does seem pretty wasteful.
Meanwhile, as a friend, I've consistently had to coach people through doubting what they've received as medical guidance from ChatGPT and other chatbot or search LLM's, variously pointing out how: (a) the response doesn't actually agree with the cited sources nor correct information, (b) the cited are poor authorities (blogspam) and are not correct themselves, or (c) both.
Thankfully, the consequences in both these kinds of scenarios have been innocuous to date, so ostensibly "not negative", because most ChatGPT/chatbot inquiries are really for moot trivia rather than anything of true consequence, but it's repeatedly shown itself to be a pretty risky and unreliable tool, as things go, so I accept that it's only a matter of time before the true "negative experience" comes and nudge people away from it when I can.
The new "I looked up my symptoms on WebMD and it said I have cancer"?
Claude is HORRIBLE at properly representing it's sources. It just makes shit up.
The worse thing is the companies behind those tools know it, they know it’s not reliable and it looks like they want to build a world where a majority of people won’t check sources anymore and blindly trust the LLM as authority. I fear that, by a tour de force of some kind, they’ll shield themselves from giving any source is the future in the name of « intelligence ». What will happen in the next generations where young that have grown up only with LLM giving them answers they saw as truth and older generations died? What a different world it will be.
...in fact it made me realize that I'm the spark bearer, meant to guide and awaken everyone else
Who could have guessed ?
Just like social media bigshots don't allow their kids to use them, tech people using the oldest Thinkpad they can find running debian or a drug dealer doesn't consume his own product.
I made this point and a coworker asked if I had been abused. With friends like these...
Reminded of Arthur C. Clarke's Travel by Wire!
I’m American, currently living in a rural southern state.
I live near a city with a large African-American population. I’ve worked in blue color jobs. I’ve met quite a few people that might be called racist.
I’ve never heard the term ‘moon cricket’.
That’s cause it’s ancient and most of the media that would have mentioned it hasn’t been shown in polite society for decades
That’s like 1930s era slang
I have, but only because I’ve made it a point to learn about American history through contemporary sources.
For that matter, I live near the largest remaining KKK organization in the country; as you can imagine, I was exposed to a ton of racist language as a child. That’s a big part of why I spent so much time in early adulthood do historical research. I could probably list more than a dozen racial epithets for Black people that I’ve heard first-hand, but that one in particular isn’t among them.
Polk County, TN... beautiful area, it's shame that's it ruined by racists.
Nope - northern Arkansas.
This article isn't much better than the genre of "Some rando on Twitter said" style articles. They're interviewing AI raters, of which there are millions in the world. It's not difficult to find a half dozen in any group that large to espouse any position. Some kind of polling would at least provide some statistical rigour but just "we managed to find a few people who espouse this position" is there purely to reinforce people's pre-existing biases.
I found the article frustrating for the lack of examples of bad things AI does that they are avoiding. They say one avoids talking about palestinians but a lot of humans including myself do that as it's so easy to offend people. Also that they may not have censored all bad words and make mistakes occasionally but that applies to pretty much all media.
Not saying that AI doesn't have issues but the article makes a very vague case.
I noticed one of my engineers blindly trusts whatever AI says. When his teammates challenged him and showed actual data, he just went, “Oh yeah, the AI was wrong.”