Have you played with OpenAI's ChatGPT?

[Lost User]

Kornfeld Eliyahu Peter wrote:

What kind of questions had you (or would) ask?

Will your existence make our kids lazy and dependent? Kids Google all and learn too little. That's a real problem, and ChatGPT will make it worse.

Bastard Programmer from Hell :suss: "If you just follow the bacon Eddy, wherever it leads you, then you won't have to think about politics." -- Some Bell.

[Slacker007]

Kornfeld Eliyahu Peter wrote:

Have you played with OpenAI's ChatGPT?

yes. has lots of potential. I'm sure it will be much better in the next couple of years, if not sooner.

[raddevus]

I had a similar experience. It is interesting because chatGPT acts just like a dev with little experience: it just spews an answer immediately -- even though it is wrong. :rolleyes: I tried out ChatGPT and asked it to simply: how can I make tabbed content using bootstrap? [see getbootstrap.com - CSS HTML designs] First it created a solution with an old version of bootstrap. I asked it, "why did you use an old version of the library?" And then it replied and created a solution that contained an non-existent version of Bootstrap. I was completely puzzled by that. Why / How did it create a link to a non-existent bootstrap code -- used a CDN link that didn't exist? :confused: I told the AI that the version didn't exist and it attempted to create sample code with a version that does indeed exist. But this time it stopped in the middle of typing the example. ??? I explained that it was not a complete example & it apologized and did create sample code. however, when I attempted to run the code (HTML & CSS) the browser couldn't even render it. This is a fairly simple thing to do. (You can see a working sample at this jsfiddle if you like[^].) I created that sample long ago. I then attempted to direct the AI by saying, "Check out the code at <link> to help you solve the problem." But in the end it couldn't even do this simple thing. I'm sure in a year or maybe less it will be able to do it.

[jschell]

Randor wrote:

It's pretty clear that the technology is going to change the world

If I only had a dollar for every time I have heard that. Google 'changed the world' but it doesn't mean much unless you can figure out the correct question to ask in the first place.

[Lost User]

jschell wrote:

If I only had a dollar for every time I have heard that. Google 'changed the world' but it doesn't mean much

I'm glad you mentioned that. I've been heavily using search engines this week because I built myself a new workstation. I'm not exaggerating when I say that some of the things I am searching on Google/Bing this week show completely irrelevant results. If I search for software engineering terms that have the same words as music/art/books/commercial products search engines just throw everything at me that include the buzz words. Interestingly ChatGPT seems to understand exactly what I am asking about. Although I admit the information isn't always correct. This technology is brand new, I see alot of room for improvement.

[jmaida]

I have use ChatGPT. I asked for the history of AI and received some new and very interesting information. I asked for the high/low/average temperatures of Moscow for a year. ChatGPT: The average high and low temperatures in Moscow vary significantly throughout the year. Here is a general overview of the average temperatures in Moscow, based on data from the National Oceanic and Atmospheric Administration (NOAA): January: Average high of 23°F (-5°C), average low of 12°F (-11°C) February: Average high of 25°F (-4°C), average low of 14°F (-10°C) March: Average high of 35°F (2°C), average low of 23°F (-5°C) April: Average high of 51°F (11°C), average low of 36°F (2°C) May: Average high of 64°F (18°C), average low of 48°F (9°C) June: Average high of 73°F (23°C), average low of 57°F (14°C) July: Average high of 76°F (24°C), average low of 60°F (16°C) August: Average high of 73°F (23°C), average low of 58°F (14°C) September: Average high of 62°F (17°C), average low of 47°F (8°C) October: Average high of 50°F (10°C), average low of 37°F (3°C) November: Average high of 37°F (3°C), average low of 25°F (-4°C) December: Average high of 26°F (-3°C), average low of 15°F (-9°C) As you can see, the temperatures in Moscow can vary significantly throughout the year, with cold, snowy winters and relatively mild summers. The city also experiences a large range of daily temperatures, with highs often reaching into the 60s and 70s during the summer and lows dropping below freezing during the winter. I hope this information is helpful. Let me know if you have any other questions. Very cool, both figuratively and literally.

"A little time, a little trouble, your better day" Badfinger

[Lost User]

Sound beep in c64 assembly: 476,000 results Sound beep in c64 basic: 1,330,000 results Pattern recognition. Spam filtering. Weighing. Language translators. 20-40 minute reads with examples. The tools are there for anybody.

"Before entering on an understanding, I have meditated for a long time, and have foreseen what might happen. It is not genius which reveals to me suddenly, secretly, what I have to say or to do in a circumstance unexpected by other people; it is reflection, it is meditation." - Napoleon I

[Lost User]

Quote:

Moscow experiences a wide range of average high and low temperatures throughout the year. Based on information from the National Oceanic and Atmospheric Administration (NOAA), the following basic summary of Moscow's average temperatures is provided:

[Spinbot - Article Spinning, Text Rewriting, Content Creation Tool.](https://spinbot.com/)

"Before entering on an understanding, I have meditated for a long time, and have foreseen what might happen. It is not genius which reveals to me suddenly, secretly, what I have to say or to do in a circumstance unexpected by other people; it is reflection, it is meditation." - Napoleon I

[jmaida]

ChatGPT draws on a lot sources. No surprise.

"A little time, a little trouble, your better day" Badfinger

[RainHat]

I like the was this totally stops the machine. 1. Stop all interrupts. 2. Enter into an infinite loop. Worst code possible.

[DerekT P]

Yes, a little. In some cases the answers seem very good, in others they can be incomplete, contradictory, or just wrong. When errors are pointed out, ChatGPT sometimes improves the answer (and sometimes apologises), but it seems to me it generally bases its responses on the first thing it looks up, rather than comparing a number of answers and finding a consensus. When I asked "whats the difference between walschaerts and stevenson valve gear?" (types of valve gear used on steam locomotives) it gave a clear answer, describing not just the differences but the advantages/disadvantages and why you might adopt one over the other. It felt like talking to a knowledgeable engineer with good communication skills. But when I asked "how do vacuum brakes work on a railway train" the initial answer was poorly described and included some confused and misleading descriptions. When I asked it to clarify an aspect it had pretty much fudged, it gave me an equally poorly worded response that said pretty much the opposite of what it had just said. I challenged it on its contradiction and it apologised, admitted it was wrong and gave a third poorly-worded spiel that didn't clarify anything and still used wrong concepts and omitted key aspects. As with (for example) newspapers, we take information and tend to assume it's true. But when it relates to a field we're knowledgeable about we quickly spot the flaws and inconsistencies and have to assume that the source is just as uninformed about other subject matters, and that we're being fed complete nonsense a LOT of the time.

Telegraph marker posts ... nothing to do with IT Phasmid email discussion group ... also nothing to do with IT Beekeeping and honey site ... still nothing to do with IT

[Nelek]

DerekT-P wrote:

As with (for example) newspapers, we take information and tend to assume it's true. But when it relates to a field we're knowledgeable about we quickly spot the flaws and inconsistencies and have to assume that the source is just as uninformed about other subject matters, and that we're being fed complete nonsense a LOT of the time.

Sadly many many people doesn't see it. And it gets even worse, when the one who explains bullshit is someone famous in social media.

M.D.V. ;) If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about? Help me to understand what I'm saying, and I'll explain it better to you Rating helpful answers is nice, but saying thanks can be even nicer.

[Kornfeld Eliyahu Peter]

The interesting thing about this chat AI, is how it evolves... At the end of our discussion I gave it a nice solution (actually the BELL form the C128, a well proved code). The AI acknowledged that it is a good solution (based on what?), but two days after still offering silly code when I ask the same question...

"Everybody is a genius. But if you judge a fish by its ability to climb a tree, it will live its whole life believing that it is stupid." ― Albert Einstein

[Nelviticus]

I believe the point of ChatGPT is not to give correct answers but to sound like the answers have been written by a real person. The fact that it's sometimes also correct is just luck.

Regards Nelviticus

[MonkeyVaughan]

I'm a software engineer, mostly contract stuff so I like to keep learning the latest technologies and stuff. I'm completely blown away by ChatGPT's possibilities. I don't feel like it will put me out of a job but it will make my work life easier, for example when I was testing what it was capable of I got it to write me a C# .Net MAUI, cross platform mobile speech to text interface and implementation, RESTful WebAPI calls and text to speech interface for responses, including dependency injection, unit tests and saving conversation snippet history to a SQLite database. Pretty blooming impressive! Some of the code it produced isn't perfect but it is a shed load better than google and even StackOverflow. A large part of our job as software engineers is knowing how to ask the right questions and being able to sort the wheat from the chaff but watching it go creating about 3 or 4 days work in about 20 minutes that saved me a lot of typing really blew my mind. To show my friends and family that I'm not completely obsessed by the whole "The future has arrived!!" vibe I was giving off I've also got it to write very funny screenplays for scenes of their favourite TV shows such as Friends, Charlie's Angels, The Big Bang Theory incorporating them in the storyline and the screenplays it produced were astonishing, kept really well within the style of each show and their characters. You can give it a theme/style for a chef's menu and it can produce one all formatted and provide recipes. It is only limited by your imagination for asking the right questions and giving it the inputs to come up with answers. It's a bit scary but amazing.

[Lost User]

MonkeyVaughan wrote:

It's a bit scary but amazing.

The only thing I find scary about it is that it can mimic humans and go undetected. I think ChatGPT looks useful for determining what a human is asking about. In other words, I see an immediate use for search engines and Q/A websites.

[Damien Hansen]

I think that AI-powered language models like ChatGPT can be used to assist with tasks such as language translation and language learning, which can make it easier for people to communicate and access information from around the world. :thumbsup: