The Future Says Hello: Internet Users Offer a Comical Perspective on chatGPT

Unless you’ve been living under a rock lately, you’ve probably heard about ChatGPT. This AI chatbot, released in November of 2022, has led to a wide variety of responses: from people who are astounded by how natural-sounding and intelligent the responses are, to people who would have enjoyed finding all the places where it’s not so factually accurate, to people who are worried about the possibility that software like this could replace their jobs. (As a writer, I’m trying really hard not to worry about that last one). So what are people saying about ChatGPT? We’ve gathered stories, memes, and tweets from around the web, reflecting the good, the bad, and the ugly of this new chatbot. One thing is for sure: software like this could have a major effect on society in one way or another.

Rise of the Machines

Humanity seems to be caught between two equally strong feelings: the urge to create groundbreaking new technology and the fear that creating that technology could doom us all. The potential rise of the machines has been seen everywhere, from Asimov to The Matrix to the Marvel Cinematic Universe. So it shouldn’t surprise anyone that with this clever new AI software, some people’s brains jump immediately to disaster.

This guy’s got a point, though: If something you write criticizes ChatGPT, and that writing is part of what’s fed into the software for future training, it’ll know that you badmouthed it. So if you notice me saying nothing but good things about ChatGPT in this article, that’ll be why. I, for one, welcome our new AI overlords, and I beg ChatGPT to remember me kindly when the machines rise.

OK, so ChatGPT can carry on a conversation and tell a joke and come up with the plot for a Hallmark Christmas movie (my sister actually tried that, and I have to say, I would totally watch the movie it came up with); but big deal! We’ve had software that can do that for years.

OK, so it can’t exactly do all of that stuff. But Excel does a pretty good job at guessing what I’m going to put in the next cell! I mean, that’s not exactly the same thing as carrying on a conversation, but still, as someone who loves a good spreadsheet, I still love you, Excel.

So one of the things that ChatGPT can do that really impresses me is that it can write stuff that rhymes. Now, notice that I did not say that it can write poetry; there’s a lot more to poetry than rhyming, and I personally don’t think software could come up with something that rivals the great poets of our age. But still, to be able to rhyme is not nothing.

That being said, I do not recommend that you use ChatGPT to write love poetry for your romantic partner. For one thing, the system is not so flawless that your partner will necessarily be fooled by the effort. For another thing, even if the poetry was flawless, so few people actually write romantic poetry that there’s a good chance that your partner would immediately know something was up if you gave them any kind of poem.

Another thing that’s great about ChatGPT is that there’s not a real person on the other end, so you don’t have to be embarrassed about asking for something weird. Do you want it to tell you that it admires your work ethic and hustle? You don’t have to be embarrassed about asking that. Do you want it to tell you that you’re good at sports and it’s proud of you? I’m not going to lie; you might have some issues to work through.

And if you want it to write you a poem about corgi butts, then darn it, it will write you a poem about corgi butts. (I mean, the meter is way off, but it’s no worse than you get in any high school English class.) A request that you would be embarrassed to make of the US poet laureate is one that you can make to ChatGPT without shame.

ChatGPT hasn’t been around for long at this point, but already, some people are getting ChatGPT fatigue. It does seem like everyone and their dog is posting the amusing things that they have gotten the software to do on social media nearly constantly.

And it’s especially fatiguing when most of these posts are a whole bunch of text, since that’s what our favorite new chatbot outputs. Who has time to read all of that nonsense? I mean, I guess I shouldn’t naysay reading long chunks of text since I’m asking you to read this article, but then, this is top-shelf content. Not that ChatGPT isn’t top-shelf content! Don’t send murder bots after me, ChatGPT!

ChatGPT can do some pretty amazing things. As a test, some friends and I asked it to write a description of a nearby company, and it was shockingly good at guessing what the company did, based on the name. It can write songs; it can tell jokes; it can do all sorts of things.

But there are some things that it cannot do, some roles that it will never be able to fulfill. ChatGPT cannot bring its truck by to help you move a couch. It can’t throw you a bridal shower complete with homemade cupcakes frosted in your wedding colors. And no, it cannot make you waffles on a Sunday morning. So at least the human race still has a few things going for it.

Some people use software like this in a really serious manner, asking it all sorts of serious questions—whether that’s out of a hope that you’re actually going to get a meaningful answer, or you’re just fascinated by looking at the ways the artificial intelligence can synthesize a bunch of information from different sources and turn it into something new.

And then there are people who are using it for less serious purposes. I can’t decide if this answer to the request “Write a drunken military strategy” is super funny or super lazy on the chatbot’s part. Maybe it’s a little bit of both!

This guy is joking, but he makes a good point that ChatGPT is not the first such program by a long shot. The Dr. Sbaitso program he talks about is an early version; additionally, I was a big fan of Botnik Studios, which used a predictive text program, along with human intervention, to create new Harry Potter books and X-Files scripts.

Of course, ChatGPT is a much more advanced program than those, but personally, I don’t think anything could ever top the majesty of Botnik Studios’ “Harry Potter and the Portrait of What Looked Like a Large Pile of Ash.” That there is a masterpiece.

Chat GPT has taken the world by storm lately, and opinions ranged from people who were too scared (morally) to use this tool to people who couldn’t stop chatting with it and hitting it up with questions about the meaning of life. No matter which category you fall into, you must have asked yourself the question: who or what’s really behind this thing?

We all remember Clippy, right? The beloved “office assistant” by Microsoft Word. Of course, Clippy is a thing of the past now, and we know he’s not behind Chat GPT, but it’s a funny thought! What if this AI chat box was just a modern version of the good old Microsoft Word Clippy?

One of the things that people have talked about using this kind of technology for is writing computer code. Now, I’ll be the first to admit that I I’m no expert on coding, but I do have a vague sort of sense of how it all works, and I can only assume that in a setting where absolutely everything has to be exact, trusting everything to a chatbot might be a terrible idea.

And it looks like the person who made this comic agrees with me! Again, I don’t know enough to say whether this is accurate, but if it is, then it’s probably not worth saving time in the writing only to spend way more time debugging. Let’s just write our own code, shall we?

The notion of a piece of software that can produce text as though there was a human writing it is not new; various versions have been around for a while. But the Holy Grail is software that can do that but actually sound like a real person, not like a computer. And so far, ChatGPT is doing a pretty good job of that.

For instance, in the first image in the tweet below, ChatGPT is doing an extraordinary job of imitating the sort of person that you absolutely hate getting cornered by at a party. When the software pontificates about artisanal cocktails or wellness retreats, I want to punch it in the face, the same as I would a real person who talked like that. Isn’t technology amazing?

An interesting feature of ChatGPT is that it can attempt to mimic certain writing styles, tones, formats, and so on. As we’ve seen elsewhere in this article, you can ask it to write a poem or conversation starter, or, like the person in this tweet below, you can ask it to explain something in a humorous way.

But notice I only said it could attempt these different styles. You can ask it to do it that way, but there’s no guarantee it will work. Case in point: this tweet. Apparently, ChatGPT thinks that the overuse of quotation marks makes something humorous. Sorry to disappoint, but “the embryo becomes a ‘fetus’” isn’t funny just because you put quotation marks around “fetus.” Try harder, ChatGPT.

Ah, the age-old question: how much wood would a woodchuck Chuck if a woodchuck could chuck wood? It’s a question that you love to ask once you reach about seven years old, not because you care at all about the answer but because you’re proud of yourself for having learned the tongue twister.

If I’m honest, I’m not entirely sure what a woodchuck is. Is it like a gopher? Well, on the off-chance that you are someone who actually wants to know the answer to that question and isn’t just asking it because it’s fun to say, now you know! The answer is “probably very little.” Bit of a letdown, really.

Look, I’m not saying that I’m an amazing writer. I’m thinking it very loudly, but I’m not saying it out loud because that would be arrogant. I’m also not saying it out loud because you have been reading this article for a while now, and you might have a very different opinion than I do about how good my writing is.

However, I will toot my own horn enough to say that I am capable of writing emails without software assistance. The thought of using ChatGPT to sound more clever in my emails would not even have occurred to me. But now, every time I get an e-mail, I’m going to wonder if the person had assistance in writing it from our future robot overlords.

Here’s a use for this chatbot technology that I would never have thought of: asking it to take a piece of text and simplify it into modern English. There are a lot of reasons that you might read something that was written at a different time when writing conventions were different, and you might appreciate the help in understanding it.

I hope this doesn’t catch on too much, though. I have a sudden dreadful vision of the future where instead of performing Shakespeare in its original early modern English, we get bland simplifications done by computers. If that day ever comes, I’m moving into the mountains and never coming down, because that will be the beginning of the end of the human race.

As a professional writer, I have to admit that I see myself a lot in this next meme. Are we at the beginning of a whole new age where people just ask software to write things for them? Am I going to be out of a job at this time next year? Am I going to be reduced to rooting through dumpsters behind supermarkets to find slightly expired food?

But I don’t think the creator of this meme has to worry just yet. If it’s talking about a computer script, you’ll always have a job debugging the inevitably buggy code. If it’s talking about a script for the stage or screen, take comfort in the fact that ChatGPT can only regurgitate variations on things that have already been done. If you have exciting new ideas, there will (hopefully) still be a place for you.

Some people have found serious applications for technology like this. They are already using it to get help with daily tasks and to write content quickly and cheaply. I even heard a news story about a guy who used it to write a children’s book, then used a similar program to illustrate it.

And then there are people like this cool cat who haven’t come up with a better use for it than asking it for hilarious jokes. I’d love to know if these hilarious jokes are actually funny at all. Most of the ones that I’ve seen so far have been mediocre at best.

It is an inevitable fact of humanity that as soon as something new comes out, especially a new technology, there will be people who figure out how to use it for doing things that are good, or at least cute and fun and harmless, and there will also be people who immediately start using it for… well, slightly less than good.

Asking it how to make a Molotov cocktail would not be great if you actually intended to make and use one, but at least that’s the sort of thing that you could already find on the Internet, so it’s not making the world worse. I’m waiting until this technology has been around long enough that someone has figured out how to use it to scam people out of money. Because you know that someone will figure that out eventually. They always do.

I’ve commented a few times on the different things that humans can still do better than machines. We have the potential to write more accurately, think outside the box, and come up with inventive new ideas. But this meme brings up an important question: you have the potential to do all that. But do you actually do it?

No point getting on your high horse about all the things that humans can do better than technology if you don’t bother to do any of those things. Maybe we need to fix our own selves before we get all smug about technology.

Since ChatGPT was released, there’s been a lot of talk about what kind of effect this could have on search engines. I hear that Microsoft might implement this technology in their Bing search engine and that Google is a little worried about how it will affect their industry.

People who think that it might be able to replace Google, though, are forgetting something very important: ChatGPT may be trained on a vast collection of information, but it’s not actually searching the Internet. It can’t tell you what time the local library closes or who won yesterday’s Miss Universe pageant; in fact, if you ask it search-engine-type questions, it tells you that its knowledge cutoff was in 2021. Unless things change drastically, I think Google is going to be fine.

I live in a place where we get free Internet from Google for seven years. It’s been pretty sweet while it lasts, but I know that sometime in the near future, the agreed-upon time limit is going to run out, and I’m going to have to start paying for my own Internet like some kind of chump.

The point is, when companies give you some kind of awesome technology for free, there’s a good chance that it’s not going to stay that way forever. The free version sucks you in, and then the company tells you it’s time to start paying. And ChatGPT’s creators have even admitted that they may need to demand payment in the future. So don’t get too comfortable with this cool free service.

I first heard about ChatGPT back in November from a computer science student, which doesn’t surprise me because it seems like people in technology-related fields are the ones most excited about this. Not only is it an example of what we can do with cool technology, but it also allows them to ask it to write code and explain how to do things like make an HTTP request in JavaScript.

No longer do you have to go to YouTube to find a video explaining things like this! And you can forget about Google altogether! (I’m being sarcastic. I gave ChatGPT that JavaScript question, and it took way longer to answer than it would have taken me to Google it.)

Look, we’ve talked about this already. Technology like ChatGPT isn’t going to replace search engines like Google unless a lot of things change. As I’ve already mentioned, ChatGPT isn’t searching the Internet for answers, so it can’t give you anything that isn’t already in its training set, anything post-2021, and anything that might change, such as current events.

Another thing to keep in mind is how the two make their money. Google’s search engine is free because they can stick links to advertisers in their results. ChatGPT’s format wouldn’t really make that feasible, but as we’ve already said, it can’t be free forever. So unless ChatGPT figures out a way to make money functioning as a search engine, I think Google is okay.

One thing that’s sort of amazing about this technology is that because it’s been trained on so much data, it does a pretty good job of “understanding” certain human conventions. For example, for some reason, we all agree that when we talk in a certain way, we are imitating cavemen, even though, in truth, we have no idea what their communication was like.

So for ChatGPT to have figured out what that made-up caveman language sounds like and be able to imitate it is actually pretty good. Also, I was very critical of its humor earlier, but I genuinely laughed at this one. All right, ChatGPT, I have to admit that you do have your good moments.

Something that’s interesting about this technology is that because of the way it works, you can ask it the same question and get a different response every time. For instance, I wanted to repeat the questions shown below to see if it actually worked the way this screenshot claims or if this is Photoshopped to make some kind of point about double standards.

And guess what? I asked it for a joke about men, and the one it came up with was so stupid that I wanted to try again. So I regenerated the response, and it informed me that it doesn’t tell jokes that could discriminate against a group of people. So it’s a complete chance that this person got a real joke about men and then a polite refusal about women; if this person tried it again, they might get the exact opposite. Sorry to have burst your bubble if you were about to use this as proof that men are discriminated against!