# ChatGPT: Have you tried it? **URL:** \ **Category:** Miscellaneous and Personal Stuff I Must Share\ **Tags:** ai\ **Created:** [January 20, 2023, 3:19pm UTC](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512 "2023-01-20T15:19:00Z")\ **Posts on this page:** 20\ **Page:** 1
**Author:** ![XOldiesJock](https://avatars.discourse-cdn.com/v4/letter/x/ec9cab/32.png) [@XOldiesJock](https://boards.straightdope.com/u/XOldiesJock)\ **Post date:** [January 20, 2023, 3:19pm UTC](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512/1 "2023-01-20T15:19:00Z")
I’d been reading about ChatGPT, an AI writing tool that apparently has become quite popular. I’ve used creative writing my entire professional career, but have rarely dabbled in fiction, and I was especially intrigued by its ability to write short stories. So, I gave it a one-sentence assignment to write a short story about the misadventures of our two cats, one young and mischievous and the other older and wiser, the sort of story a young child might enjoy. It did, in about 30 seconds. The story was charming, with a believable situation, touches of humor and an uplifting conclusion. Honestly, it brought a tear to my eye. With a little rewriting and embellishment, I wouldn’t be embarrassed to submit the story to a publisher of fiction for children. I can even vividly imagine what the artwork would look like. I got to wondering, though, if this would constitute plagiarism. I gave it a simple idea, the characters and their motivation, but ChatGPT came up with the plot. Does anyone have any thoughts on that? And have you tried ChatGPT for anything, and what was the result? ---
**Author:** ![scabpicker](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/scabpicker/32/8268_2.png) [@scabpicker](https://boards.straightdope.com/u/scabpicker)\ **Post date:** [January 20, 2023, 3:52pm UTC](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512/2 "2023-01-20T15:52:17Z")
Ehh, who owns AI generated art is kind of up in the air at the moment, I believe. I can report that it sucks at writing lyrics. A friend of mine and I spent an afternoon trying to get it to write good ones after reading the article below. It writes lyrics that might be acceptable if they were written by an elementary school kid, but still pretty terrible. So they’re laughably bad, and it has a tendency to name-check the author you’ve asked it to emulate in the chorus as if it thinks everyone writes like Biggie Smalls or something. > **[Nick Cave says ChatGPT's AI attempt to write Nick Cave lyrics 'sucks'](https://www.bbc.com/news/entertainment-arts-64302944)** > > The musician responds after a popular chatbot produces lyrics "in the style of Nick Cave". ---
**Author:** ![BeepKillBeep](https://avatars.discourse-cdn.com/v4/letter/b/9e8a1a/32.png) [@BeepKillBeep](https://boards.straightdope.com/u/BeepKillBeep)\ **Post date:** [January 20, 2023, 4:02pm UTC](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512/3 "2023-01-20T16:02:04Z")
Just FYI, there’s a long thread on ChatGPT you might be interested in. I’ve tried it from the point of view as an AI scientist to test it, and try to understand its inner workings. It is pretty cool, while somewhat overhyped. ---
**Author:** ![Si\_Amigo](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/si_amigo/32/2877_2.png) [@Si\_Amigo](https://boards.straightdope.com/u/Si_Amigo)\ **Post date:** [January 20, 2023, 4:05pm UTC](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512/4 "2023-01-20T16:05:36Z")
I showed it to my coworkers yesterday in a team meeting and asked it do an example of filling out our weekly reports to make them more interesting (subject being very boring cost reduction activity). The site is down right now, but its response made everyone laugh and then they admitted that there were phrases that it came up with that they actually would incorporate into a weekly report. ---
**Author:** ![pulykamell](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/pulykamell/32/3166_2.png) [@pulykamell](https://boards.straightdope.com/u/pulykamell)\ **Post date:** [January 20, 2023, 4:12pm UTC](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512/5 "2023-01-20T16:12:42Z")
It’s equally terrible at poetry. It tries, that it does, but it’s terrible. That said, I had absolutely zero expectation of it being good at that. Even its meagre efforts surprised me. It’s great – for me – at brainstorming. I’ve fed it bits and pieces of stuff I wrote 20 years ago that were sitting in a notebook somewhere (I’m sure plenty of us have nuggets of creative writing we’ve just abandoned), and it’s been fantastic at helping flesh out characters, giving numerous possible scenerios for where a story should go, to specifically adding fantastical characters I hadn’t even considered, and keeping things cohesive. As long as you interact with it as a buddy you’re bouncing ideas off of, I think it’s nothing short of a magic. There have even been times I’ve fed it a small part of a finished story and asked it how one could proceed, and it had a very good idea of my intent and even played around the same themes and ideas I was generating in the beginning I fed it. Pretty much came to the same ending, though with much plainer and less interesting language and dialogue. So that’s what I’ve found fun with it. I’ve spent hours just chatting about characters, various plots twists, asking it to try making more absurd suggestions, and it’s been a great catalyst for my imagination. For me, it’s not overhyped in the least. I’m continually astounded and surprised by it. ---
**Author:** ![Paul\_was\_in\_Saudi](https://avatars.discourse-cdn.com/v4/letter/p/b77776/32.png) [@Paul\_was\_in\_Saudi](https://boards.straightdope.com/u/Paul_was_in_Saudi)\ **Post date:** [January 20, 2023, 5:15pm UTC](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512/6 "2023-01-20T17:15:07Z")
My wife watches knitting videos on YouTube. In one of them, the host asked ChatGPT to answer several fairly technical questions sent in by viewers. (“How does the material of the needles effect your knitting?”) I was quite impressed by the replies, even if the host pointed out a few mistakes. If the program can do that on a fairly obscure subject now, imagine what it might do on a more mainstream topic in a few years’ time. ---
**Author:** ![Ruken](https://avatars.discourse-cdn.com/v4/letter/r/f475e1/32.png) [@Ruken](https://boards.straightdope.com/u/Ruken)\ **Post date:** [January 20, 2023, 6:43pm UTC](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512/7 "2023-01-20T18:43:00Z")
I tried generating rap lyrics on various science topics. They were not good. ---
**Author:** ![Darren\_Garrison](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/darren_garrison/32/92_2.png) [@Darren\_Garrison](https://boards.straightdope.com/u/Darren_Garrison)\ **Post date:** [January 21, 2023, 2:14am UTC](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512/8 "2023-01-21T02:14:58Z")
> [@XOldiesJock](#): > > Does anyone have any thoughts on that? And have you tried ChatGPT for anything, and what was the result? > [@The next page in the book of AI evolution is here, powered by GPT 3.5, and I am very, nay, extremely impressed](https://boards.straightdope.com/t/the-next-page-in-the-book-of-ai-evolution-is-here-powered-by-gpt-3-5-and-i-am-very-nay-extremely-impressed/975945/): > > Hi dopers. I used the word page in the topic deliberately because this thread is about an amazing text-based AI. You can try it yourself by making a free account at [chat.openai.com](https://chat.openai.com/chat). Here are some of my prompts. You can type normal english and make spelling mistakes and the AI will still understand you just fine! Write the first page of a gripping scifi novel in Asimov’s style. The robot stood at the edge of the cliff, gazing out at the vast expanse of the Martian landscape. It was the lates… ---
**Author:** ![Ponderoid](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/ponderoid/32/19221_2.png) [@Ponderoid](https://boards.straightdope.com/u/Ponderoid)\ **Post date:** [January 21, 2023, 2:20am UTC](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512/9 "2023-01-21T02:20:50Z")
> [@XOldiesJock](#): > > With a little rewriting and embellishment, I wouldn’t be embarrassed to submit the story to a publisher of fiction for children. I can even vividly imagine what the artwork would look like. > **[This man used AI to write and illustrate a children's book in one weekend. He...](https://www.businessinsider.com/chatgpt-midjourney-ai-write-illustrate-childrens-book-one-weekend-alice-2023-1)** > > ChatGPT and Midjourney helped Ammaar Reshi make what he thought was a fun gift for his friends' children. He was shocked by the response. For the TL;DR types, Midjourney is an AI art generator. ---
**Author:** ![Sam\_Stone](https://avatars.discourse-cdn.com/v4/letter/s/ecccb3/32.png) [@Sam\_Stone](https://boards.straightdope.com/u/Sam_Stone)\ **Post date:** [January 21, 2023, 2:45am UTC](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512/10 "2023-01-21T02:45:57Z")
The quality of its output highly depends on the quality of the input. There are many tricks you can use to improve the output, which people are learning as they go. For example, I gave it this prompt: “Please write a scene of humorous dialog between a narcissist and a psychopath, who are arguing over what to do with a cat they just rescued.” ChatGPT: > [@](#): > > Narcissist: “I think we should keep the cat. I mean, look at it, it’s absolutely adorable. Plus, having a pet will do wonders for my image. Everyone will see how caring and compassionate I am.” > > Psychopath: “Are you kidding me? We can’t keep that thing. It’s going to be a nuisance, always getting in the way and making noise. We should just let it go.” > > Narcissist: “Let it go? Are you crazy? This cat is a symbol of my kindness and generosity. I can’t just let it go.” > > Psychopath: “I couldn’t care less about your image. All I know is that this cat is going to be a problem. We should just get rid of it.” > > Narcissist: “Get rid of it? No way. We’re keeping this cat and that’s final.” > > Psychopath: “Fine. But when it starts scratching up your designer furniture, don’t come crying to me.” > > Narcissist: “Don’t worry, I’ll make sure it only scratches the cheap stuff. Now, let’s go post a picture of our new pet on social media. I can already feel the likes pouring in.” > > Psychopath: “Whatever, just as long as it stays out of my way.” Interesting but fairly pedestrian. The way to improve it is to give more details to ChatGPT, which can be done through conversation. So I asked ChatGPT to describe a narcissist, then a psychopath, then I said, “Given what you just told me, write a scene about two people nmed Nora and Pete arguing over a stray kitten.” ChatGPT: > [@](#): > > Nora: “I saw the cutest little stray kitten on my way home today. I think we should take it in.” > > Pete: “Why would we do that? It’s just another mouth to feed and another responsibility to deal with.” > > Nora: “But think about how cute it would be to have a little kitten around the house. Plus, it would be a great conversation starter when we have guests over.” > > Pete: “I couldn’t care less about what our guests think. And I don’t want to be bothered with another living thing in the house. It’s all about me, not some stray kitten.” > > Nora: “You’re so selfish, Pete. Can’t you see how much joy a little kitten could bring us? Plus, it would be a great addition to my image, as I’ll be seen as a savior of the innocent.” > > Pete: “I don’t need a kitten to boost my image. And I don’t want to waste my time and resources on something as insignificant as a stray kitten. I have better things to do with my time and money.” > > Nora: “You’re impossible. I’m going to go get that kitten, with or without your help.” > > Pete: “Suit yourself. But don’t expect me to lift a finger to take care of it.” The first conversation is very on-the-nose, using pretty much simple definitions of narcissists and Psychopaths. But by getting ChatGPT to first generate a list of features for both, the input becomes much more complex, resulting in better output. Next I asked ChatGPT to give me a description of an apartment a typical poor person might live in in a big city. It did so. Then I said, “Okay, now rewrite the argument as a scene in such an apartment. write it like you are a great writer.” ChatGPT: > [@](#): > > The small apartment was dimly lit, the only source of light coming from a single window that was partially obscured by a dirty curtain. The air was thick with the smell of damp and mildew, and the floorboards creaked beneath the weight of the couple’s footsteps. > > Nora stood in the center of the room, her arms crossed tightly over her chest. “I saw the cutest little stray kitten on my way home today,” she said. “I think we should take it in.” > > Pete, who was lounging on the threadbare couch, let out a sigh. “Why would we do that? It’s just another mouth to feed and another responsibility to deal with.” > > Nora’s eyes widened in frustration. “But think about how cute it would be to have a little kitten around the house,” she pleaded. “Plus, it would be a great conversation starter when we have guests over.” > > Pete shook his head. “I couldn’t care less about what our guests think. And I don’t want to be bothered with another living thing in the house. It’s all about me, not some stray kitten.” > > Nora’s frustration turned to anger. “You’re so selfish, Pete,” she spat. “Can’t you see how much joy a little kitten could bring us? Plus, it would be a great addition to my image, as I’ll be seen as a savior of the innocent.” > > Pete let out a bitter laugh. “I don’t need a kitten to boost my image. And I don’t want to waste my time and resources on something as insignificant as a stray kitten. I have better things to do with my time and money.” > > The tension in the room was palpable as Nora glared at Pete, her hands balled into fists at her sides. “You’re impossible,” she said through gritted teeth. “I’m going to go get that kitten, with or without your help.” > > Pete shrugged indifferently. “Suit yourself. But don’t expect me to lift a finger to take care of it.” > > As Nora stormed out of the apartment, the door slamming shut behind her, Pete lay back on the couch and closed his eyes. He didn’t care about the kitten, but he couldn’t shake the feeling that something vital was missing from his life. ChatGPT can handle 8K of token in its input. That’s roughly 5,000 words or so. The easiest way to generate complex inouts is tomstart a new session, then ask questions that cause ChatGPT to generate lots of details, then feed that in woth your prompt by telling ChatGPT to write its output after considering everything it’s told you. ---
**Author:** ![Moonrise](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/moonrise/32/14764_2.png) [@Moonrise](https://boards.straightdope.com/u/Moonrise)\ **Post date:** [January 21, 2023, 10:09am UTC](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512/11 "2023-01-21T10:09:29Z")
I was surprised at how judgmental it could be. I dabbled, and was rather impressed, with it when I fed it reasonable requests (e.g. give a definition of linguistic typology in the form of a play by Samuel Beckett). But when I asked it to give me a recipe for Mac & Cheese written by a drunk Shakespeare, I got stuck in a loop where it would basically answer variations of “it is unethical to depict real persons as drunk”. ---
**Author:** ![xtenkfarpl](https://avatars.discourse-cdn.com/v4/letter/x/c2a13f/32.png) [@xtenkfarpl](https://boards.straightdope.com/u/xtenkfarpl)\ **Post date:** [January 21, 2023, 4:05pm UTC](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512/12 "2023-01-21T16:05:07Z")
Darn thing has been “too busy” every time I’ve tried to access it? ---
**Author:** ![Hari\_Seldon](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/hari_seldon/32/5173_2.png) [@Hari\_Seldon](https://boards.straightdope.com/u/Hari_Seldon)\ **Post date:** [January 21, 2023, 8:12pm UTC](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512/13 "2023-01-21T20:12:34Z")
I couldn’t get on either. My son asked it a question, "Who discovered the [a mathematical construction] and it totally fanned on it. The construction was perhaps a bit obscure, but there was a specific published paper that described it. ---
**Author:** ![carnivorousplant](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/carnivorousplant/32/3563_2.png) [@carnivorousplant](https://boards.straightdope.com/u/carnivorousplant)\ **Post date:** [January 21, 2023, 8:20pm UTC](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512/14 "2023-01-21T20:20:32Z")
Start with, “There is another system.” ---
**Author:** ![Sam\_Stone](https://avatars.discourse-cdn.com/v4/letter/s/ecccb3/32.png) [@Sam\_Stone](https://boards.straightdope.com/u/Sam_Stone)\ **Post date:** [January 21, 2023, 10:22pm UTC](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512/15 "2023-01-21T22:22:07Z")
There are already about 5 or 6 LLMs out there doing what ChatGPT does. I just read that it cost about $50 million in computer time to train ChatGPT. That’s a lot of money for a startup, but not a lot for a large tech company that wants to get into the LLM game. And the cost will come down over time, Moore’s law slowdown notwithstanding. These things are going to be everywhere. And through APIs we will eventually find an LLM lurking behind a whole bunch of stuff you wouldn’t expect. ChatGPT also announced their paid model - $45 per month if you want API access, guaranteed 24/7 availability and faster output. Otherwise, the free one will remain available but will probably be even harder to get on as they prioritize paid subscribers. I saw a cool plug-in description - it’s a plugin for a word processor or E-mail, where you can right-click on a sentence and ask for five alternatives if you’re not happy with the one you wrote. You can also just right-click wherever you are and choose , “please finish this for me” and ChatGPT will read what you wrote, figure out style and subject, and finish your E-mail or dpcument section for you. We are going to see stuff like this everywhere. There are a lot of people out there who are very talented in their field but terrible writers due to dyslexia or other issues, and in the digital age this holds them back. For them, ChatGPT will be a great equalizer in E-mail communications, letters, etc. ---
**Author:** ![scabpicker](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/scabpicker/32/8268_2.png) [@scabpicker](https://boards.straightdope.com/u/scabpicker)\ **Post date:** [January 22, 2023, 12:06am UTC](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512/16 "2023-01-22T00:06:07Z")
> [@carnivorousplant](#): > > Start with, “There is another system.” > [@Sam\_Stone](#): > > There are already about 5 or 6 LLMs out there doing what ChatGPT does. I’m pretty sure that was a reference to Colossus: The Forbin Project > **[Colossus: The Forbin Project](https://en.wikipedia.org/wiki/Colossus:_The_Forbin_Project)** > > Colossus: The Forbin Project (promoted as The Forbin Project) is a 1970 American science-fiction thriller film from Universal Pictures, produced by Stanley Chase, directed by Joseph Sargent, that stars Eric Braeden, Susan Clark, Gordon Pinsent, and William Schallert. It is based upon the 1966 science-fiction novel Colossus by Dennis Feltham Jones. > The film is about an advanced American defense system, named Colossus, becoming sentient. After being handed full control, Colossus' draconian logic... ---
**Author:** ![carnivorousplant](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/carnivorousplant/32/3563_2.png) [@carnivorousplant](https://boards.straightdope.com/u/carnivorousplant)\ **Post date:** [January 22, 2023, 1:27pm UTC](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512/17 "2023-01-22T13:27:56Z")
Indeed. ---
**Author:** ![Chronos](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/chronos/32/134_2.png) [@Chronos](https://boards.straightdope.com/u/Chronos)\ **Post date:** [January 22, 2023, 11:51pm UTC](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512/18 "2023-01-22T23:51:25Z")
And of course, even if it’s expensive to train one of these things, it’s cheap to take one that’s already trained and make a bazillion copies of it. It’d pretty much just be the cost of the hard drive space. A quick Google isn’t revealing just how big the trained model is, but the training data was somewhat less than a terabyte, and the trained model is almost certainly smaller than that. ---
**Author:** ![pulykamell](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/pulykamell/32/3166_2.png) [@pulykamell](https://boards.straightdope.com/u/pulykamell)\ **Post date:** [January 23, 2023, 12:04am UTC](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512/19 "2023-01-23T00:04:12Z")
Geez. And what’s even a terabyte these days? (Granted, that’s text.) I don’t know how many terabytes of hard drives I have lying around. At least 40 or 50. (My current main data drive is 4TB, over 50% full, and I just put it in in August, to replace the previously filled 4TB drive). ---
**Author:** ![Sam\_Stone](https://avatars.discourse-cdn.com/v4/letter/s/ecccb3/32.png) [@Sam\_Stone](https://boards.straightdope.com/u/Sam_Stone)\ **Post date:** [January 23, 2023, 2:15am UTC](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512/20 "2023-01-23T02:15:05Z")
ChatGPT has about 175 billion parameters. That’s its size. it would be that size if it read a gigabyte of data or a petabyte. A parameter is just a number in a matrix or a vector. So, a 32-bit parameter would need four bytes of memory, so 700 billion bytes, or 700 gigabytes. My understanding is that the entire model has to be in RAM to be efficient. So it’s not going to be running on a PC. Smaller models can. Here’s an article describing how to run a 6.7 billion parameter LLM on a PC with 12GB of RAM. This would normally take 54 GB, but the author splits the model to be partly in CPU RAM, partly in GPU RAM, and the rest cached to the hard drive. > **[Run Very Large Language Models on Your Computer](https://pub.towardsai.net/run-very-large-language-models-on-your-computer-390dd33838bb)** > > With PyTorch and Hugging Face’s device\_map I wonder how long it takes to process an input. This stuff is heavily math intensive. [Next page](https://boards.straightdope.com/t/chatgpt-have-you-tried-it/978512.md?page=2)