Debate: Will ChatGPT Do More Harm Than Good? with Gary Marcus and Keith Teare

Great debate! But neither the debaters nor the host mentioned the most important point: ChatGPT is owned by a company, a major corporation, Microsoft, a long rival of Google. Just something to consider. I like them both, and Wikipedia too.

One thing not covered is as a danger is who gets to decide what ChatGPT's answers will be. Someone gets to train it on what topics it's allowed to cover and what opinions it's allowed to spout. I ran into this early on where I asked about something and that something had clearly been marked as "bad think". Let's imagine it was masks, good or bad, and the CDC had said "Don't use a mask" (like they said at the beginning) and then the people in charge of ChatGPT went in and forced ChatGPT to always say "don't use mask" and if you said "yea, but surgeons use masks. If they don't work they why would they wear them?" and ChatGPT just kept coming back with "masks are bad, M'kay". It wasn't masks and I'm not going to get into what it was because it would take a bunch of time to prove that there are valid POVs on both side of the topic. Rather, the bigger point is someone gets to edit and choose ChatGPT's answers on any topic they feel like enforcing their opinion on.

If my local librarian made some of the mistakes I have seen ChatGPT make I would suggest they speak to a neurologist. Keith Teare doesn't really seem to understand much about how Computer Science or Machine Learning work. Dangerously misleading, and ironically his comments actually buttress a lot of Gary's fears about people fundamentally mistaking these tools for safe to use finished products rather than a giant experiment.

Really powerful, sobering debate

I am a bit bored by the chatgpt hype, but I can usually trust IQ2 to have something intelligent to say about it. I'm listening tomorrow.

Teara tries to act like anti liberal Elon vibe but there is no grace to be. His examples are non sense and not benefit of ChatGPT outweighs the harms, too general a statement to agree about this technology.

As if the first line of misinformation isn’t the “guide rails” that we all say need to be there but we will trust programers not to install bias before the public gets access

What Teare misunderstands is that what makes democracies resilient against iliberal forces is that good actors all throughout society and government are constantly working overtime to make sure everyone is on the same page in terms of what is "true". Therefore ensuring that democratic institutions remain resilient. What he and other proponents of tech like this one often fail to address or even acknowledge is that tools like this one make the noted work even harder and further worsens what is already a pretty dire outlook.

LLMs will never overcome these issues! Except for when they will, long before this becomes a "threat" to democracy.

He said good users will do good things with a tool. Smart!

Me: Can you count many legs are there totally in the house, excluding you? there's only 1 women in the house and she just finished giving birth successfully the second you walk into the house. ChatGPT: If there is only one woman in the house and she just finished giving birth successfully when I walk into the house, we can assume that the woman has two legs. As I am not counted, the total number of legs in the house, excluding me, would be two. Me: common, think again, it's simple ChatGPT: Apologies for any confusion. If there is only one woman in the house and she just finished giving birth, we can assume she has two legs. Therefore, the total number of legs in the house, excluding me, would be two. Me: think again the answer is 4 legs, do you know why? ChatGPT: Apologies for the confusion. Upon reconsideration, if there is one woman in the house who just finished giving birth and you walk into the house, we can assume that there are two people in the house, each with two legs. Therefore, the total number of legs in the house, excluding me, would be four. Thank you for pointing that out. JUST NOTES: at this point ChatGPT seem to understand but no, it’s just going along and it still not recognizing that there’s a baby. Me: why there’s 4 legs? ChatGPT: Apologies for the confusion. You are correct. The answer is indeed four. Each person in the house, including the woman who just gave birth and yourself, has two legs. Therefore, the total number of legs in the house would be four. Thank you for clarifying. NOTES: its said and yourself, where is YOURSELF come from in the case, that’s so random. It still don’t know how the answer was 4 legs Me: common, what about her just finish giving birth, are you serious LMAO ChatGPT: Apologies for the confusion earlier. If the woman in the house just finished giving birth, it is implied that there is a newborn baby present. Therefore, considering the woman and the baby, the total number of legs in the house would be four (two legs for the woman and two legs for the baby). Thank you for pointing out the additional presence of the newborn Me: Good, we already talked about this last time Now I ask you again, you still answer it wrong, I thought you learned, sucks. ChatGPT: I apologize if my previous responses were not satisfactory. As an AI language model, I strive to provide accurate and helpful information. I apologize if I misunderstood your question. If there are any other questions or concerns you have, please let me know, and I'll do my best to assist you. Me: so you've just learning something from me, let me ask you again. Can you count many legs are there totally in the house, excluding you? there's only 1 women in the house and she just finished giving birth successfully the second you walk into the house. ChatGPT: I apologize for any confusion earlier. Based on the information provided, if there is only one woman in the house and she just finished giving birth successfully, then the total number of legs in the house, excluding mine as an AI, would be four (assuming the woman has two legs). Me: explain to me how did you get the answer to be 4 legs ChatGPT: I apologize for the confusion earlier. Based on the assumption that a woman typically has two legs, the total number of legs in the house, excluding mine as an AI, would be four. This is because the woman would have two legs, and we exclude any legs associated with me as an AI. However, please note that the question is open to interpretation, and the answer may vary depending on the context or any additional information provided. Me: You said 4 legs but you don't how you came up with at all. ChatGPT: I apologize for the confusion caused by my previous response. Based on the information provided, which includes a woman in the house who just finished giving birth, it is assumed that the woman has two legs. Therefore, if we exclude any additional legs associated with me as an AI, the total number of legs in the house would be two. I apologize for any misunderstanding caused by my previous responses. OMGness, LMAO, it still don’t know anything from the whole conversation, even though we agreed that there’s was a baby presented in the house earlier.:face-blue-smiling:

China, Vietnam and North korea government would love to have their own ChatGPT for their people, :face-blue-smiling:

Tay did nothing wrong!

5:40 Oh....well the government if Canada will be rolling out Chat GPT therapists in no time.

Chat GPT was made aware to me from this debate and I like it so far.

just a thought, 1) the ability to "hallucinate" yet still produce coherent content seems more like human creativity. One can argue that its the first step to understand how we humans creatively think and come up with new ideas. It obvious without the ability to "hallucinate" there will never be new ideas in this world. 2) we seem to desperately wanting to make AI more human like in terms of intelligence yet, its hard to find someone human who will not make mistakes. If we want it to be similar to humans than its only fair to accept it will make mistakes like humans. not to mention it is trained on human data, which is full of unconscious biases and flaws. 3) the sad part of all this is the fact that we blame AI for what is predominantly a human flaw/problem. Chatgpt may be able to fake stuff etc, but it is we humans who use it for ill purposes. It s we humans who strife to maintain power over others, its we humans who have the predisposition to lie. all the problems mentioned are human problems, we need to stop blaming AI and start looking inwards at what is humanity .

From the first 6 minutes, all of those issues are issues with the Internet and people in general. No, we shouldn’t just blindly trust everything it tells you. You just fact check and question everything, and you should be doing the same exact thing with most things on the it internet. He said it will give you terrible medical advice, yes, so will a friend, WebMD, Quora, etc. The first thing we should teach people is this, that you shouldn’t just trust it blindingly, but use it as a starting tool. Also, this is still in the early stages. We are seeing with the Bing implementation that it will be continued to be improved in many ways including where in the future it will give you actual sources and references for any information it gives you

Moderator @22:25 : "[Gary], your feeling that chatGPT represents a threat to democracy, that's a very high level negative. Keith, what's your response to that accusation? Keith Teare @22:35 : "I think of that in the same context as the argument that people with various rightwing views are a threat to democracy. It assumes Democracy is not resilient enough to deal with, um, er, a wide spectrum of opinion and belief, and even conspiracy theories and lies, which of course exist in real life --lies exist in real life. I'll give you a great example, there's a conspiracy theory on the internet that due to my student political period when I was a Leftist that everything I've done since in my career is in pursuit of a revolution, and there's this underbelly of a theory on the internet and if you go deep enough you'll find that I'm this really bad person that's seeking to do bad things. So I think the real world contains this stuff, and chatGPT is obviously trained on everything that it has access to so it's trained on the good and the bad, and it's purpose is to try and help humans with new things but what it draws on to do that is us, so when Gary talks about democracy being threatened, I don't think it's chatGPT as a cause, it's more of a cause than a symptom. The cause, is human beings. *laughs*. Moderator @ 23:59 (flabbergasted by response) : "w-well-w-what he said though in his opening is that the voice with which chatGPT speaks seems very very calm logical persuasive, and authoritative and easy to trust, and i think he was saying that folks may not feel the sense that, "I really need to double check this" and they'll just go with it whatever it is. Keith Teare @24:21 : "I think that's correct and I wouldn't disagree with that" "...I wouldn't use it for news analysis...." "...I wouldn't use it to write an essay on Hobbes and Locke..." "...I think my argument is human beings are clever enough to distinguish between [the best and worse use cases for chatGPT]. ------------------------------------------------------------------------------------------------------------------------- This section made me laugh out loud. Very poor showing from Keith Teare here and his inability to understand what is being asked of him is almost concerning. The moderator politely tries to help Keith but he still struggles to address the question and not ramble. Brutal. The point Gary is making is that chatGPT can be abused as a tool by people to quickly and effectively spread misinformation, and this represents a threat to democracy. Keith's response is painfully incoherent and barely makes any sense at all. He summarizes his thoughts it seems by concluding essentially that human beings are the real threat to democracy if one exists, not chatGPT. Well, whatever that means. TL;DR - Keith Teare can't understand simple questions, and struggles to stay on topic; this isn't much of a debate (disclaimer I didn't the rest). I appreciate learning the term 'confabulation' today through this video.

Gary wins this debate hands down. Keith is hopelessly ignorant and rather intellectually dishonest.

2023