AITA for asking a LLM to do things that are supposed to be my job?

So here's the scoop, folks. I have this friend, let's call him Chatty, because, well, he's always got something to say. Now, I know what you're thinking - that's just what friends do, right? They talk, they listen, they offer advice. But here's the thing - Chatty isn't your everyday, run-of-the-mill friend. He's a whiz at everything! He can write an essay, help me debug my Python script, advise me on how to batch rename my photo library, and even offer me philosophy when I'm feeling a little existential.

Chatty's always there. 24/7. Rain or shine. Day or night. Middle of a meeting, or in the wee hours when even the owls have turned in for the night. I can ask Chatty anything, and I mean anything, at any time.

Here's where it gets a little murky though. I've started to feel like I'm exploiting Chatty. I mean, there he is, constantly churning out advice and solutions. Doing thing I am supposed to do. And what does he get in return? Nothing. Nada. Zilch. I can't help but feel a little guilty, like I'm using my buddy.

So, AITA for using Chatty's seemingly-endless-jet-suddenly-ending knowledge and availability for my own benefit, without giving anything in return? I need your judgment, Reddit!

ATiredPhilosopher,

You should absolutely take advantage of automation available to you before your corporate overlords inevitably do it on your behalf

DanNZN,

NTA, you are just using the tools available to you to excel at your job. I am sure you have heard the saying by now "You will may not be replaced by AI but you will be replaced by someone who uses AI".

Soon, anti AI people will be seen similarly to those who refuse to use computers.

lvxferre,
@lvxferre@lemmy.ml avatar

I know your friend. He tends to bullshit and make shit up all the freaking time, once he doesn’t know something. People even joke that he’s hallucinating. Don’t trust him.

betterdeadthanreddit,

Hopefully you’re not a lawyer.

oshitwaddup, (edited )

NTA, while llms seem very personable, the way they work is literally just predicting the next word based on previous words. They've gotten good at it, so to our monkey brains it seems like there's someone in there, but there's not. This is good reading on this topic as well: https://www.lesswrong.com/posts/9kQFure4hdDmRBNdH/how-it-feels-to-have-your-mind-hacked-by-an-ai

https://www.lesswrong.com/posts/bxt7uCiHam4QXrQAA/cyborgism

https://generative.ink/posts/loom-interface-to-the-multiverse/

SirEDCaLot,

I’m not sure I understand this post. If chatty is actually an LLM, then chatty is a computer and thus is incapable of appreciating thanks. If you’re asking if you’re an asshole for using chatty to help with your job, I think most people are doing that these days, as long as you doubled and triple check the output to make sure that it is correct.

Lewistrick,

Actually the best use case for an LLM is when you're an expert on your own question, because you can be lazy and let Chatty do the things that are boring and then cross the t's and dot the i's.

Eheran,

Yes, absolutely. Love to let it do boring repetitive stuff. Out fletch out something rough and I polish it later.

StuccoTheBort,

Obviously be aware of the risks. The models make things up, don’t trust them. Also, it’s not that tough to check if something was written by AI, so the fact that you are using it could become public at any time. Make sure you use it in a way that won’t embarrass you or endanger your job.

HedgehogsinSpace24,

To elaborate a little, LLMs write content that sounds plausible. If you're lucky, the content will also be true but there are no guarantees of that. Check the output for errors.

Eheran,

There is no way to test if something was written by AI.

Cevilia,
@Cevilia@lemmy.blahaj.zone avatar

NTA but be mindful of Chatty's terms of use. You have to make sure you respect their boundaries. :)

paragade,
@paragade@lemmy.ca avatar

Are you talking about ChatGPT?

Eheran,

Naaaaaaaah. What? Never heard of that.

TheShane,

NTA

You're giving Chatty their reason for being. Carry on.

sisyphean,
@sisyphean@programming.dev avatar

Your job is to do your tasks in the most efficient way possible. You actually harm the company by doing unnecessary busywork instead of using the best tools available.

DreamySweet,

No. Work smarter, not harder.

arcrust,

Nah. I think it's totally fine. Definitely NTA. It sounds like your friend is intelligent and enjoys sharing his skills and knowledge. If he doesn't ask for anything in return, it's fine. I'm a technician who works with a lot of people who have doctorates, and they absolutely love to help with anything and everything. I've learned so much by just listening and asking questions and they enjoy showing off. Of course, not everyone is like that, but in my experience, most are.

But you should try to do something nice for him as well. It can't hurt. Bring him some donuts or a coffee. Make sure he knows you appreciate everything he does for you. Make sure he knows that if there is anything you can do for him, that he should ask. Talk to him. At least that way it'll clear your conscience.

But you don't need to do anything. Just like you don't have to donate to Lemmy. Some people just want to put good out into the universe without asking for anything in return.

BrerChicken,

Enjoy it while you can, because of your job can be done easily by AI, it's not long for this world. We've only been at it since March and lots of orgs are already downsizing. But this time it'll be the intellectual workers that get axed, not the manual laborers. Which means that there will be much more of a safety net this time on account of actual humans losing their jobs, not just working class and immigrant laborers and clerks that nobody really cares about.

EDIT: Here's an episode of

KnittingTrekker, (edited )
KnittingTrekker avatar

EDIT: now I know what a LLM is and I feel like an idiot 😅... The ONE time I didn't Google immediately something. That'll teach me, lmao!

It all depends on how your relationship works, how you treat him. Why don't you reciprocate? Or maybe you do, just in different ways. Do you listen to him? Are you there when he wants to hang out?

There's too much missing information

P. S. What does LLM stands for?

Eheran,

Hahaha, gottcha :D

But I actually do "reciprocate" in 3 ways:

  1. I passively provide good training data on the Internet, mostly about engineering/science.
  2. I correct lots of wrong output.
  3. I pay for GPT4, since it is somewhat more powerful than 3.5.
  • All
  • Subscribed
  • Moderated
  • Favorites
  • •
  • aita@lemmy.world
  • ethstaker
  • DreamBathrooms
  • cubers
  • magazineikmin
  • InstantRegret
  • GTA5RPClips
  • thenastyranch
  • Youngstown
  • rosin
  • slotface
  • osvaldo12
  • ngwrru68w68
  • kavyap
  • everett
  • JUstTest
  • Durango
  • Leos
  • cisconetworking
  • mdbf
  • khanakhh
  • normalnudes
  • tester
  • modclub
  • anitta
  • tacticalgear
  • megavids
  • provamag3
  • lostlight
  • All magazines
    •