This has been a hot topic of late (on the forum and in the news), so I’m creating this thread as a centralized place to discuss false or misleading information concerns for any of the various AI search/chat engines.
Also moved a few of the previous posts on this topic here, to keep things organized.
Great story about Turley (the attorney - I don’t know if I would call him prominent as the article does) being accused of sexual harassment by ChatGPT. ChatGPT reportedly made up sexual harassment allegations against a prominent lawyer (msn.com)
I think he’s “prominent” as he is a regular legal commentator for FoxNews (they seem to have a very short list) but something is really fishy - when they tried to replicate the bot request, it refused. Had it already been retuned or was the list made up for other purposes?
I am trying something here because I believe this is an important article. NYT says I can “gift” ten articles per month so hopefully this works.
The gist of this is that AI is not ready for prime time; many internally in Microsoft and Google have strongly opposed it on ethical and technological grounds; and the heads of MS and Google are casting all caution to the winds in the race to be the dominate force in the “next big tech.”
Thanks for scaling the paywall for us. 
I recognize the importance of the issue, but we still need to keep things organized on the forum (especially with AI, which naturally sprawls into a bevy of topics). The article talks vaguely about broader societal concerns, but I felt “societal concerns” would be too general a topic name.
As is stands, I think misinformation captures most of the immediate concern around AI; and for the broader issues, we can create new topics as they arise and become more clear. How does that sound?
I’m fine however you characterize it - the real concern is that this is very much like Nadella and Pichai are racing to be Pandora without any concern for the consequences…
Has there ever been a point in history, where a nation has laid down a weapon; or a company has delayed a competitive edge? No, I think Nadella, Pichai, and all on them, don’t have a choice. (And if they ever did choose caution, you know that the next ‘visionary’ CEO is just waiting to grab their spot. 
)
Relevant article on the Why of the “confabulations”.
Or conflagrations…
Reads like another fluff piece.
Someone, some time around now, was going to release an ‘AI’ no matter the consequences. And there’s nothing really anyone could have done about it. So Microsoft just saw this inevitability and decided to try and get the first mover advantage amongst the tech behemoths.
It’s here, and lamenting over if we should and or shouldn’t have let it loose (which ultimately that article, and many others are doing - just with a veneer of the blatantly obvious ‘caution is needed’) is a waste of time and energy.
We are still very early days on all of this, but we and a couple of our customers have found at least one useful application of these tools (and that’s really all they are IMHO) which is in the area of data mining.
One customer who is a market research/presence firm, has been getting a great deal of useful information on perceptions and knowledge levels of various brands and products in the market simply with the query of “tell me what you know in a few paragraphs about…”
It of course needs to be moderated and filtered by informed humans, but it surfaces both obscure real facts, but also common misperceptions about a product or brand.
And for us, it’s been useful as a possible entry tool with queries along the lines of "tell me about the companies producing shock and water-resistant tablet like devices…
Apologies in advance, but that is just a cop out. They knew it was not ready for prime time, but they launched without concern for the consequences - look how Microsoft reduced its own ethics team for overseeing AI. Sometimes the almighty profit cannot be the only goalpost.
Ask it to tell you about the remaining companies with plans to use pen silos with their phones, tablets, and laptops…
It’s not a cop out.
Someone was going to do it and be the leader, and if not some tech giant that can need shamed/financially issued into some sort of responsibility, then some body with fewer/no qualms.
And quite frankly, what are you, I, or anyone going to do be able to do about it?
I’ll just keep b******g everywhere I have a voice on the web. If I had any real clout I’d join the open letter by Wozniak and company…
Sorry to sound like a frightened little child, but until this year I never thought I’d see this in my lifetime, and NEVER considered how it would have to be trained to get to this point - it is ingesting the entire digital record of humankind - guts, feathers, and all.
I totally agree with this sentiment. AI has no true sense of self-awareness and it never will (sorry Data fans of Star Trek: The Next Generation, but this is one area that Star Wars got right). This gives AI a distinct and dangerous disadvantage compared to thinking and feeling human beings because AI in many queries must implicitly make moral decisions while lacking any real capability to do so. There are too many in society who have “god syndromes” like the CEOs of these tech companies. These people fail to realize that AI must be treated like a powerful tool. Any tool can become a weapon when put into the wrong hands which, in turn, can be weaponized against others and even backfire against its operators.
So the thread is already getting a bit apocalyptic 
Though remarkably no mention of Skynet (until just now 
)
Sadly, I think it’s going to take a major tragedy, directly, provably caused by AI driven software, before there is any real attempt to reign this in.
Of course gun control advocates keep thinking that after every mass shooting too 
Well, AI is more like an airplane on full autopilot with the added risk of no clear end destination specified to the operator until it has finished execution. Its variability is one aspect that makes it highly dangerous.
A little test done with bing Chat GPT:
(Q):
Does the triangle A = (10, 0) B = (10, 10) C = (9, 5) And the triangle D = (10, 0) E = (9, 5) F = (10, 10)
are the same?
(A):
Two triangles are considered congruent if they have the same shape and size. In order to determine if two triangles are congruent, we can use several methods such as SSS (side-side-side),
SAS (side-angle-side), ASA (angle-side-angle), AAS (angle-angle-side), HL (hypotenuse-leg) or RHS (right-angle-hypotenuse-side). In this case, we can see that triangles ABC and DEF are not congruent because they have different side lengths and angles.
(Q):
Are the vertices identical?
(A):
Two triangles are identical if there is a triangle correspondence so that corresponding sides and angles of each triangle are equal in measurement. In this case, triangles ABC and DEF are not identical because they have different side lengths and angles.
(Q):
Are the vertices of the triangles identical in pairs?
(A):
Two triangles are identical if there is a triangle correspondence so that corresponding sides and angles of each triangle are equal in measurement. If two triangles have identical vertices, it means that they have the same set of vertices. However, if they do not have the same side lengths and angles, they are not identical.
(Q):
Are A and D the same?
Are B and F the same?
Are C and E the same?
(A):
Triangle ABC and DEF are not identical because they have different side lengths and angles. Therefore, A and D are not the same, B and F are not the same, and C and E are not the same.
(Q):
Thanks for your help.
(A):
You’re welcome! If you have any more questions, feel free to ask.
Not saying this is inaccurate - in fact I’m more concerned they are right:
