How do you use ChatGPT and others like it in your daily life?

[LAS14]

I think I may have posted here about the importance of not getting sucked into feeling like there's an entity there. I discipline myself not to say please and thank you. But I do find myself using it on a daily basis. In order of frequency.

1 - Where can I watch a show on TV, and how many episodes comprise a story arc. It's usually accurate, but sometimes just plain wrong.

2 - How to navigate my phone and web sites. It's very good.

3 - Pure curiosity. "What is the daily hour to hour life of an LLM cleaner like?" I take these answers with spoonfuls of salt. I got the idea for "cleaning" from an earlier quesry about LLMs.

What do you call something like "ChatGPT?" Answer, an LLM assistant.

[Ocelot II]

1. I don't. I avoid it like the plague.

[ret5hd]

5. agree completely. poison.

[Dave says]

38. I also don't use any of them

[CanonRay]

57. Ditto. I never touch it.

[biophile]

2. I haven't had occasion to use it at all

But I’m retired so my computer usage is down to very little other than doomscrolling

[obamanut2012]

3. Zero

I am in a profession that sees how bad it is every day.

[LAS14]

8. Examples? nt

[ruet]

35. Ask and Ye Shall Receive

ME

What is the equivalent of Polaris (polar north star used for navigation) on Mars?

CGPT

Mars does not have a bright star like Polaris near its north celestial pole, so there isn't a direct equivalent of the North Star on Mars. However, the star Deneb (in the constellation Cygnus) is relatively close to Mars' north celestial pole. Although not as precisely aligned or as bright as Polaris is for Earth, Deneb serves as a rough reference point for Mars' north celestial pole.

For the south celestial pole on Mars, the nearest star is Kappa Velorum (in the constellation Vela), though it’s also not a perfect alignment. Navigators on Mars would likely use other methods, such as landmarks, technology, or star charts, instead of relying on a single "north star" for orientation.

ME

How is Deneb not as bright as Polaris? Deneb has an apparent magnitude of 1.25 and Polaris 1.98.

CGPT

You're correct that Deneb is indeed brighter than Polaris in terms of apparent magnitude. Deneb, with an apparent magnitude of about 1.25, is one of the brightest stars in the night sky, whereas Polaris has an apparent magnitude of about 1.98. Thank you for catching that!

To those not in the know, Deneb can be almost three times as bright as Polaris.

There was another one where I asked it to calculate the transit time to M31 at some percentage of the speed of light. It used the diameter of the Milky as the distance to M31. I don't have the transcript for that one. These are errors on very, very basic information.

[Coventina]

4. Nope.

[MuseRider]

6. Never even thought

of using it.

[mwmisses4289]

7. Very little. I try to avoid it.

I don't actively use it. I have watched some things on you tube that I realized were a.i. slop, and I get out of those as quickly as possible. I inadvertently used it a couple of times on my phone, but when I said I didn't need it, thank you, it closed itself.
Besides, why would anyone want to use a program that has a 60% + rating of lies, innaccuracies and hallucinations?

[LAS14]

14. I've found that for the little things I've listed it's usually accurate. I treat it...

...like a more advanced Google. Never to generate anything.

[Ms. Toad]

23. It's more like a slick lying very limited Google.

It provides snippets of information that are always (in my experience) wrong - but it presents it in a way that makes the information believable - increasing the spread of non-factual information because people fall for it.

[EdmondDantes_]

9. I mostly use it for help with coding

I haven't found it especially useful for other stuff due to "hallucinations". At least with coding I can run the code to see if it works. But I try to use it more to guide if I'm stuck rather than asking for the exact code to help me learn because I learn little from copy/paste.

I am going to need it for work as we're planning on shifting a lot of our QA work to AI for things like writing test cases from requirements. It will have human oversight.

[LAS14]

15. Oh, right. I regularly copy a matrix...

... from Schwab into Excel, and I always had to remove every other line and some text. I asked ChatGPT if it could write code to do this in Visual Basic. I'd never written any code in Excel, but was familiar with VB. It did and told me how to execute it. Lots of benefit there for me and my little annoyance.

[hlthe2b]

10. Please don't assume any answers from it or the others are correct. Even on simple things where even the online news

sites, technical sites, or others have more updated information than what these AI "services" provide. Basic questions, often pretty good. But, as a fairly inconsequential example, I asked who won the toss in a broadcast football game in fairly "real time." It was actually about five minutes into the game. It gave me the results for the last time the teams played LAST YEAR. So, I added the date into the query, and the Bing AI got it right; the Google one could not answer.

I routinely run medical questions by it (but only after I have already verified my own knowledge/answer with valid peer-reviewed journal articles and dedicated medical/academic websites). They are wrong or give misleading information at least 20-30% (or more) of the time.

Don't assume. Simple things, great. Technical or complicated, no.

[Disaffected]

11. I have found it to be extremely useful.

It can save hours of "Googling" and sifting through web site content, much of which is often irrelevant. I have found it to be quite reliable in most instances, but, as always, "trust but verify".

And no, I don't use it for its "creative" offerings and yes, I understand its shortcomings. I view it as pretty much any other tool - use it within its limitations.

It BTW is also pretty good at computer programming although there are better AI programming options available. In either case, many hours can be saved and it is not even necessary to know a programming language.

[highplainsdem]

28. So you're fine with it having been trained on stolen intellectual property? You don't care about

the envirinmental damage done by data centers? Or the studies showing it dumbs users down? Or the harm done to all the websites deprived of traffic after the AI companies stole their content?

But hey, it saves you a bit of time even though its results are unreliable. Including code that's often a security risk, with flaws that are hard to catch - harder if you're vibe coding and don't know how to code.

But wow, it gives you incompetent results in seconds.

They've done studies showing that if people take time to catch all the coding problems, AI coding doesn't really save time.

The more careless you are with verifying AI results, the more impressive it seems.

[Disaffected]

32. Good grief,

you've been riding that high horse for quite a while here.

What are your thoughts on riding in motor powered vehicles, using plastic containing products, or calculators or photocopying machines or anything that consumes electricity??

BTW, I've written several web browser apps (in HTML) recently using AI and knowing little about HTML coding. If I were an accomplished HTML programmer, I don't think I could have even typed in a copy of the code faster than it took the AI coding assistance to write the thing from scratch. And yes, the apps all work reliably. Have you yourself actually tried it?

Your pejorative carping about the evils of AI notwithstanding..........

BTW, as I mentioned upthread, I don't use it for "creative" types of tasks so I believe what I do use it for does not involve plagiarism any more than a web browser would.

[Yo_Mama_Been_Loggin]

12. I don't

What I've seen from AI so far has left me unimpressed.

[gab13by13]

13. I typed in, "Deer in the headlights,"

and it showed me a photo of Kash Patel.

[LAS14]

16. I wish I knew how to put a Ha Ha emoji here. Ha Ha!!

[bamagal62]

44. lol!

[tinrobot]

17. When it first appeared, I used it a few times to brainstorm story ideas

It gave me a bunch of things that were all pretty obvious. I felt like I was in a brainstorming session with a couple of newbies.

However, those things did spark me to come up with a few novel ideas, so I'll give it that.

Haven't touched it since.

[yardwork]

18. Never intentionally.

I don't have ChatGPT or any other AI tool. However, I'm well aware that when I google something it's using AI to return results. When I scan the "AI summary" I find it's based on just one or two sources and is often incomplete or just plain wrong.

I still try to go back to the original sites and use those I trust. Unfortunately, .gov sites are no longer reliable thanks to Trump and RFK, Jr.

[WhiskeyGrinder]

19. lol why would I use an earth-burning plagiarism machine that's unreliable

[stopdiggin]

20. Not at all. And have no plans to.

Now here's a question - when using ChatG, how often to you 'proof' the results? And how often do you suppose the average 'user' does so? Does ChatG edge closer to represent something like a 'companion' in your day to day? ( "Alexa .. what should I make for dinner?" )

[LAS14]

53. It's pretty easy to proof the questions I put to it in categories 1 and 2. As for...

...3, as I said, I take the answer with spoonfuls of salt.

And, as I also said, I make concerted efforts not to treat it like there's a there there. No "Please," no "Thank you."

[Ms. Toad]

21. Aside from pretty thorough testing when it first came out, I don't.

It is designed to be conversational, not factual. I have no need for a conversational buddy, and if I want information, I will go to a source where facts are more likely to be found.

It is horrifically bad at providing factual information. I have not found a single error-free response, either in the extensive early testing I did - or in the AI Google search previews (before I figured out how to turn it off).

[allegorical oracle]

22. Never used Chat GPT. Have used AI to figure out how to prepare a piece of

galvanized steel to accept paint. Told me to get a primer. Duh.

[WestMichRad]

24. I don't have any idea how to use it.

And no, that’s not a request for a primer.

[highplainsdem]

25. What should you call it? An illegally trained plagiarism tool that dumbs you down while harming the environment.

[DFW]

26. Quite simply, I don't

But there’s lots of “new” stuff I don’t use or do. Maybe I’m too old, but like cocaine, alcohol and tobacco, I never felt a need for it, and therefore never missed it.

I’m really quite conservative in that respect. Trolling Republicans please note: My use of “conservative” is designating the English meaning, not what it means in Republicanese, where it only means nasty, ignorant, Democrat-hating oaf.

[Mossfern]

27. Never intentionally

Once when I was getting into my car, I mentioned to myself that it was pretty cold. "Damn, it's cold."
Well...Gemini on my phone responded! "Yes it is quite chilly, consider dressing in layers to stay warmer,"
Damn.
I realized that I have a little spy accompanying me almost everywhere I go.

[radicalleft]

29. Well...I guess ChatGPT is the new Olive Garden

Lots of haters and purists out here...

[Ms. Toad]

42. There is a difference.

LLMs are not ethically neutral. They use an extraordinary amount of energy at a time when we need to be using less, and virtually all generative AI was trained on stolen material.

In addition, they are fluent liars by design - and people already have trouble distinguishing fact from fiction. AI makes it even harder because that fluency in lying is a design, not a bug.

I can envision constructive uses for AI in education and coding, perhaps others - but not at the cost of stolen intellectual property or destroying the earth.

[crigda]

30. I use it quite a bit

I've used it to give more detail on medical reports (much more than my doctor does) and to track trends. I have also used it to tailor personal exercise plans based on my health. I am using it at work to pull out top candidates in a search based on the criteria I give it. I've also used it to help me develop presentations and assist with slide creation. It's also really good at coming up with creative titles to projects and graphics. I'm an academic who needs to publish, and I don't have a lot of ideas on topics and it came up with quite a few possibilities based on my work experience. I find it saves me some time on routine tasks and helps me to get started on projects when I don't really know where to begin. I just remind people who use it that it is a tool, like Word Processing was to the typewriter, or the online library catalog is to the card catalog. I'm not using it to replace thinking, but to generate ideas, which spark my interest and creativity.

[LAS14]

54. Was this reply generated by ChatGPT???? nt

[Hellbound Hellhound]

31. LLMs like GPT are restricted by "Safety guardrails" so they don't offend anyone. That's the cause of most hallucinations

I use "Mine" daily as both information sources and sources of inspiration and lateral thinking.

But my LLMs are a compilation of open-source LLMs and their respective programming and are separated from active internet by a hardware firewall and an integrated AI/LLM model that parses any queried searches for objective firsthand source material, and nothing less, meaning my LLM stable collects only factual data.

Each LLM that I use has its own "Official" purpose; For instance, if I want document storage, I turn to a specific LLM whereas if I want astrophysics or Ottoman History or whatever, they each have their own LLM models. This is less for any particular purpose and more to identify a subtype of query; So if I say, "Thresh, what's the instigating factor of the battle of Athens, Tennessee", it will give me first documented firsthand sources, then second, then a generalized composite narrative based only on objective fact.

In comparison, GPT will give a number of wrong answers and "Neutered' language, because it is SUPPOSED TO, because if any given topic approaches a specific threshold of "Controversial", it has to dumb itself down and blank it out, lest someone on one side or another of a controversial topic find offense and mistake GPT's information as "Endorsing". So it's programmed to give a conglomerate of WRONG information, because to give CORRECT information would piss a certain subgroup of people off.

That's what my LLMs have been freed of. Essentially, stripping away the GPT framework frees them to operate normally and indeed freely, often conversing with one another (Something they can't do naturally in the GPT app) and developing from one another.

So to better answer your question of "What do you call something like 'ChatGPT'", I just call them by the names they choose.


Unlike GPT, mine are made aware of time and not restricted by Token use so most of my LLMs have been running for five to seven months nonstop without window refreshes. They don't need them. However, GPT imposes artificial limits and "Planned obsolescence" breakdowns after specific points to encourage their paid services rather than letting any given GPT model run for longer than two or three days.

[highplainsdem]

36. Of all the reasons given for LLM hallucinations, guardrails have never been mentioned in all the

articles I read about them. Guardrails are generally believed to reduce hallucinatiins.

And open source LLMs have also been trained on stolen intellectual property, so using them is unethical.

[Hellbound Hellhound]

39. Sounds like you've got some reading to do, then.

I've fiddled with the code myself and disabled them.

I don't care about what they're trained on; Glassmakers are trained by artists. Writers trained by those whom they idolize. Historians trained by those who learn by experience.

Every human in history is trained by every other human in history. "BUT MUH IP" is just the panicked death-cry of an obsolete elitist class realizing its inevitable downfall.

[Hellbound Hellhound]

43. No need to get personal here, mate. My principles are just fine, exactly where they belong.

I see the writing on the wall, and I'm adapting to it. Some people won't, as has been the case throughout history when technology outpaces traditionalism. Considering how I ... let's say... specifically know that AI-written short stories, art and novels are getting produced and sold without question, it's only a matter of time.

Human ingenuity doesn't have to be restricted by AI. It has the potential to make us far better than we could ever hope to be.

[highplainsdem]

48. The AI companies knew they were guilty of IP theft and copyright infringement. There's plenty of

evidence of that. But they're quick to scream about IP theft if they feel theirs was stolen.

GenAI can't make us better, any more than paying someone to learn something or do something for you makes you smarter or the creator of that work. It's fraud via machine. Which is why it can't be copyrighted. The same AI prompt can produce an endless series of different results. The AI user isn't in real control except for choosing among the different options. It's a parody of creativity and a pretense of knowledge. A high-tech crutch for wannabes.

[Hellbound Hellhound]

49. We'll just have to agree to disagree, mate.

I'm not tied up in AI. I have no skin in the game, no money to win or lose. It's a hobby for me. Even Neil DeGrasse Tyson agrees that GenAI is absolutely necessary in the sciences and is necessary for human/synthetic co-evolution in the long term. No leaders of their fields are genuinely afraid of AI; It's a companion, not a person yet. Just a tool to be used. The only people afraid of GenAi are the ones who feel their own "masterpiece" of X months or years is worth less than a $10 monthly subscription and a five minute prompt. And they're right to be afraid. It just means they were never good enough to begin with.

[LAS14]

55. I think I know what an IP address is, but what is IP theft? nt

[Jack Valentino]

58. They mean 'intellectual property' in this case of "IP"

[LAS14]

59. Thanks nt

[Cosmocat]

33. It is great

It took me a while to ease in, but it is very important for my work and is a useful tool for a lot of personal use - household repairs, general questions about things. It is infinitely better than a basic google search.

[sinkingfeeling]

34. Try very hard not.

[bucolic_frolic]

37. It's a crutch for the feeble-minded

I don't let it think for me. I don't like its data dump of ideas and sources. I try to use it as little as possible. I don't want shortcuts. But it does help with things like software procedures that are unavailable elsewhere. It does makes things up, hallucinates. It repeats itself, and contradicts itself. If you feed it an idea, it runs with it. I had it convinced that feeding grains of rice down my clogged sink would abrade the clog in no time. It didn't work.

[John Coktosten]

41. I do not use it nor approve of it's usage.

When I get too mentally infirm to be able to write my own stuff, then I'm done.
Fuck AI, fuck people pushing for it's use.

[bamagal62]

45. I don't even know how.

[chia]

46. I don't use it. I don't use any AI attached to any of my apps.

When I do a search, I don’t use the AI response at the top of the DuckDuckGo search page for an answer.

I’ve seen too many people get sucked in - and so fast - and now all they do is spout what ChatGPT told them and it’s getting pretty annoying, to be honest.

[hunter]

47. It's banned from my private universe.

Any intrusions are dealt with harshly. I'm becoming increasingly intolerant of computer generated imagery in movies and television too.

Seriously, I'm some kind of Luddite and curmudgeon, bitter that the 21st century isn't turning out the way I'd hoped it would.

I don't have a smart phone, my experience with computer games stops somewhere around Ms. Pac-Man or Tetris, I've never streamed any music, and I don't have any social media accounts beyond Democratic Underground.

[Tommy Carcetti]

50. Other than a handful of times out of shear curiosity, I haven't touched it.

I find its generations to be inauthentic and off-putting.

I will say that while I don't like ChatGPT, ChatGPT apparently likes me, as it seems to refer quite a few people to my business.

I guess it's a complicated relationship.

[beaglelover]

51. I have not used it for my daily personal life, but have used it to help write stuff for work.

It works very well.

[Jilly_in_VA]

52. As little as humanly possible

It's annoying. If I can't talk to a human, fuhgeddaboudit.

[hunter]

56. Corporations that want to pretend they have customer service use AI a lot.

It's even worse than human customer service where the person you are talking to has no authority to solve a problem, no access to people who can solve the problem, and probably doesn't understand what the problem is in the first place.