AI - The Good The Bad The Ugly

[Wolkenflieger]

Net negative, I never liked it. I wouldn’t say it was “fine in the past” it was just different. At some point it was experimenting with a new tool.

Now it’s Unironically being used and takes the soul out of human interaction.

Would you rather get send a “good morning hope your doing well” text from your parents, or an AI slop image with the same words on it?

One is intentional, the other is copy paste slop.

I’d rather look at a comic sans ms paint flyer for an event or taco truck menu than an impersonal AI photo that is soulless at best, misleading and broken at worst.

Hope *you're doing well. The possessive "your" would annoy me.

 

[cali_jazz]

I still can't figure out what a normal person uses AI for but then I see people asking the AI questions and like it's a real person. Every meaningful question has an answer of "it depends" and chat bots do a piss poor job of providing that. Like if all you use for AI is data scraping a webpage, that's not even AI that's just a chat bot that automated a data import.

I use AI for all kinds of things, particularly research, gathering and organizing information, and comparing data from multiple sources. Tasks that might otherwise take me endless hours can often be accomplished in minutes. Like a curated Google search on steroids; just bring me the information and I'll figure out what to do with it or what additional questions need to be asked or areas explored.

I don't expect it to be infallible, and I certainly don't blindly accept everything it tells me. I question its conclusions and verify anything important. But dismissing AI because it doesn't save you time overlooks how useful it can be to others. It's a tool, and like any tool, its value depends on how you use it.

 

[Wolkenflieger]

I'm a pro artist and I love AI art if it's good. Artists can curate better than non-artists the same way we do with fashion or photography. Having eyes and a brain doesn't make one an artist, even with AI. This is why some people still take terrible pictures even with modern smart phone cameras.

I also write well and poor writers may assume my writing is AI, but I don't use AI for such things. The idea that someone would think my writing is AI is a downside.

AI is just a tool, like a power saw, microwave, blender, computer, airplane, etc.

 

[Wolkenflieger]

I agree with a lot of what you're saying, but I respectfully disagree on a deep point: I think we DO have the ability to change course, or stop things, or make a difference individually.

My only voice is with my money in today's society. And withholding that money from companies that I don't agree with, or more accurately, spending my money with companies that I do agree with and support, DOES make a difference. Sure not just me, but there's others like me and we can make changes. And people are the reason those industries are still alive. I choose to believe in people, and what we can do when we organize together, no matter how unpopular my optimism may be these days.

Giving up to the inevitability of anything is just that, giving up. I choose to rage against the dying of the light, to quote a poem.

I think a perfect example is the environment. I live in Huntington Beach, where the Bolsa Chica wetlands were basically a toxic waste field from the petro industry when I was a kid learning to surf and swim in Sunset Beach. But a bunch of locals banded together, and now we have a protected wetlands where I can walk and see critically endangered species nesting and feeding. That affects me personally, and was grass roots driven by less than 50 people.

Another example would be the quality of the air here in the greater Los Angeles area. The air was brown when I was a kid, and only cleared up on Sundays after the weekday traffic remnants had finally settled. Today I can see the mountains from the beach nearly every day. Not to mention I don't have chronic bronchitis like I did when I was a kid living in the urban pollutant capital of the state.

Like you said, Global Warming exists today, and we are aware of the dangers. That doesn't mean we can't act today to have meaningful changes in the future. We just feel small because we are small. But banding together our voices grow and have impact.

And I do agree that we need to keep up with threats from abroad, just like we need weapons to defend ourselves, AI is now a necessity in many ways.

And I will agree that change is inevitable, and change is uncomfortable.

We can still put meaningful guardrails and regulations in place that protect and put people and the environment first.

You can thank EV adoption for the cleaner air too. I haven't used a gas vehicle since 2020.

 

[rodralig]

I still can't figure out what a normal person uses AI for but then I see people asking the AI questions and like it's a real person. Every meaningful question has an answer of "it depends" and chat bots do a piss poor job of providing that. Like if all you use for AI is data scraping a webpage, that's not even AI that's just a chat bot that automated a data import.

I use AI for all kinds of things, particularly research, gathering and organizing information, and comparing data from multiple sources. Tasks that might otherwise take me endless hours can often be accomplished in minutes. Like a curated Google search on steroids; just bring me the information and I'll figure out what to do with it or what additional questions need to be asked or areas explored.

I don't expect it to be infallible, and I certainly don't blindly accept everything it tells me. I question its conclusions and verify anything important. But dismissing AI because it doesn't save you time overlooks how useful it can be to others. It's a tool, and like any tool, its value depends on how you use it.

Earlier this year, I brought my eldest to a police K9 charity shooting match. Nothing sanctioned like USPSA or IDPA, just a local charity event. It was my teen's first match, so naturally, I wanted to see how he did.

Apparently, the organizers didn't use PractiScore, so I reached out to them for the results. They sent me what looked like a full CSV export with everyone's scores, stages, times, penalties, etc. Problem was, I only had my phone with me.

So rather than open that massive CSV and try to make sense of everything on a 6.3-inch screen, I just copied and pasted the whole thing into GPT. Asked it to parse the data, organize it, and give me a breakdown of names, times, penalties, and overall rankings.

Took maybe 2–3 minutes, from copying the CSV to getting a nice, readable summary of the results, including how my kid did against everyone else. I'd call that a pretty good time saver, especially giving my kid that immediate smile. 55/73.

 

[Darqhelmet]

I actually prefer human design as well. Our logo, the word mark, is from a person, but like people, they can be very unreliable and took some money and left. I never got my initial set of shirts from the guy and it paid him quite a bit of money.

Also, I try to do it where I can, where it’s tasteful, and I try to mix and match where it makes sense. I always appreciate the human element to things, but people are very inconsistent, just like AI.

The reality is, I can’t afford artists for every single image that we would produce. especially in California in the Bay Area. This niche in particular is very price sensitive. We know that because the site helps people save money and get information.

The only irony is some people who save money also criticize the people who try to earn a living, saving them money, because of the tools. Things only move faster because now it’s easier to get things done, relying on people like third-party developers who charged well into the thousands for a simple fix wasn’t sustainable, and why we moved so slow in the beginning. Once I was able to invest in the equipment and level up my understanding of the technology, our deployment sped up, and a number of features implemented. Which would’ve taken a year with a third-party developer and costs well into $10K+. You can see the OG site still struggling with this

Eh, I don’t think anyone here is attacking you. Reddit, F-’’em that entire sight is a clown show.

We disagree on AI, not a big deal. Doesn’t mean I think personally you’re a jerk or evil. Just means we disagree. Everything online is amplified so disagreements seem like a way bigger deal than they are. You can’t see the person, hear them, etc.

I think I called someone a retard, and got a nono note for it. It’s text on the screen it reads worse than it is. in texts i’m a monster in person im easy to get along with.

 

[Darqhelmet]

I use AI for all kinds of things, particularly research, gathering and organizing information, and comparing data from multiple sources. Tasks that might otherwise take me endless hours can often be accomplished in minutes. Like a curated Google search on steroids; just bring me the information and I'll figure out what to do with it or what additional questions need to be asked or areas explored.

I don't expect it to be infallible, and I certainly don't blindly accept everything it tells me. I question its conclusions and verify anything important. But dismissing AI because it doesn't save you time overlooks how useful it can be to others. It's a tool, and like any tool, its value depends on how you use it.

Yeah, but how often are you getting wrong information and not even knowing it?

I’ve had it tell me wrong info on power supplies, torque spec for cars, size of hardware, etc.

“my bad, you’re totally right to push back on me…” is not acceptable. It’s not thinking, it’s not alive, it presents information as an authority when it has none. Search is worse because of it, the internet was an amazing source of information and now has been poisoned to the point where I trust nothing it says. I have gone back to manuals, tech sheets, MDS, etc. Nothing you search for anymore can be trusted.

 

[decius_mus]

Yeah, but how often are you getting wrong information and not even knowing it?

I’ve had it tell me wrong info on power supplies, torque spec for cars, size of hardware, etc.

“my bad, you’re totally right to push back on me…” is not acceptable. It’s not thinking, it’s not alive, it presents information as an authority when it has none. Search is worse because of it, the internet was an amazing source of information and now has been poisoned to the point where I trust nothing it says. I have gone back to manuals, tech sheets, MDS, etc. Nothing you search for anymore can be trusted.

When using it for research you're supposed to validate all of the information with the links it provides you. If it doesn't provide links you need to tell it to do so. If you're using ChatGPT, you can create project folders with a set of instructions that tell it what information and sources you prefer. So you can ask it to pull data from tech sheets, manuals, government websites and the like and it'll provide direct links to them so you can check them yourself. Furthermore, when using it as a research tool you're not supposed to take the output at face value, and you need to continuously push back on its assumptions and have it push back on yours until you create a working model of what you're researching.

 

[Wolkenflieger]

As far as people not thinking for themselves, I think this has already happened and is happening. In many situations, mob-rule is already the order of the day and it infects whole sites (like Reddit). Certainly, college campuses seem to be hotbeds of indoctrination too. The bubble effect is all too real. Allowing people to challenge you once in a while is healthy, and it's critical (still) to know logic, the laws of logic, and logical fallacies to avoid. The art of debate seems to be lost on the younger gen, because now those who make the wrong noises are simply canceled.

A lot of people think that they can spot AI art, but really how would they even know? I think most people just assume it because they're not used to seeing quality 3D art, and quality art takes time. What people are looking for are mistakes, but I've seen so much great 3D Art that people today would just assume is AI, and this was before AI was a thing. Non-artists can catch obvious mistakes (TOO MANY FINGERS) but they miss the things artists catch, which is why they're not artists.

The avatar in my PDP, was this created with AI? How as a non-artist can anyone even tell? In the end these are just pixels. This has been a downside for sites like ArtStation where artists inevitably started using AI to augment their art (or used AI entirely) and it sullied the real work artists had to do to "compete". A good curator can generate hundreds of great AI images in a day compared to the weeks it takes for a 3D artist to create and render something.

The advantage of course to real 3D art is permanence and control. 3D art isn't going to change from frame-to-frame, and it can be animated. Nearly every video game anyone is playing relies on quality 3D art, though I'm guessing AI is being used for some UI and other items, all curated by artists and directors.

As a 3D artist, I have to actually show my work if I were applying for a job with a portfolio. I have to show concepts, 3D wireframes, turnarounds, and sometimes animation. This is also how we could rule out theft in the days before AI was a thing, as in theft of someone's 3D-rendered images. But, this too will change. Eventually, AI art will too have permanence and detail that will look just like well-crafted 3D renders...or complete movie scenes.

Unlike most of my 3D artist colleagues, I love AI art if it's well-curated because it's just another tool, like 3D art itself and the way it replaced 2D art. That was a freaky time and I remember it well. A lot of artists simply refused to learn 3D, but some embraced and mastered it. Either way, it was scary, but nobody doubted that 3D art was here to stay even when it was in its infancy.

AI Actors are coming. Real actors will simply license their likeness for use much like musicians license songs. Imagine grabbing Tom Cruise's likeness at age 30 and using that in your own film, with obvious restrictions dictated by the license. Winona Ryder at 28. Same thing. Actors will still make their money, but soon everyone will be their own filmmaker, director, composer, editor, etc. and it'll all happen way faster than anyone can imagine. Actors will have permanence and be compelling, drawing from decades of film for expressions, look, movement, gestures, etc. You as the AI-director will craft your film as you see fit, and it will be as awesome as you can make it. It will not suffer simply because it was made with AI anymore than Toy Story suffered for being the first feature-length all-3D movie.

I'm looking forward to it, that and my Optimus robot of course.

 

[Nomad]

It's coming whether people like it or not. I love it for repetitive tasks and being more efficient with my time, but it does make mistakes and can be biased, so you better know what you're doing if you're gonna use it for work.

 

[RustVendor]

Earlier this year, I brought my eldest to a police K9 charity shooting match. Nothing sanctioned like USPSA or IDPA, just a local charity event. It was my teen's first match, so naturally, I wanted to see how he did.

Apparently, the organizers didn't use PractiScore, so I reached out to them for the results. They sent me what looked like a full CSV export with everyone's scores, stages, times, penalties, etc. Problem was, I only had my phone with me.

So rather than open that massive CSV and try to make sense of everything on a 6.3-inch screen, I just copied and pasted the whole thing into GPT. Asked it to parse the data, organize it, and give me a breakdown of names, times, penalties, and overall rankings.

Took maybe 2–3 minutes, from copying the CSV to getting a nice, readable summary of the results, including how my kid did against everyone else. I'd call that a pretty good time saver, especially giving my kid that immediate smile. 55/73.

That's not AI it just automated a couple easy Excel formulas that would take the same amount of time to apply yourself if you were familiar. No one is saying that kind of easy task automation is inherently problematic.

 

[USMC06RET]

AI makes it so much easier to drone your targets using automatic target recognition. It's supposed to reduce collateral damage but that only applies if you are deploying the correct ordnance. If the operator ends up killing the wrong target(s), he gets to blame it on the AI. Then we also have completely autonomous AI weapons which take the human out of the equation well, except the human targets.

Ya gotta love technology!

 

[Wolkenflieger]

It's coming whether people like it or not. I love it for repetitive tasks and being more efficient with my time, but it does make mistakes and can be biased, so you better know what you're doing if you're gonna use it for work.

True, and I would submit that with anything, it's important to retain one's critical-thinking skills. This is important for anything from weather to news to AI anything, and some AI are far more..."biased"...than others. Critical-thinkers will spot this and move accordingly.

 

[RustVendor]

I use AI for all kinds of things, particularly research, gathering and organizing information, and comparing data from multiple sources. Tasks that might otherwise take me endless hours can often be accomplished in minutes. Like a curated Google search on steroids; just bring me the information and I'll figure out what to do with it or what additional questions need to be asked or areas explored.

I don't expect it to be infallible, and I certainly don't blindly accept everything it tells me. I question its conclusions and verify anything important. But dismissing AI because it doesn't save you time overlooks how useful it can be to others. It's a tool, and like any tool, its value depends on how you use it.

I would also say Google is a bad database. That's the inherent issue is that the underlying data is bad because it's based on internet comments and websites where it can't differentiate between an expert or not. But also I think both can be true, that people find value in it and it saves them time but it also contributes to being another layer in between the actual data and their understanding. it also just depends what we are calling AI, is it AI because you asked a question in plain text vs a Google search? Technically you used an LLM to use plain text. Or is it only AI when it provides an answer in plain text? I think the LLM aspect is secondary to a model which produces a conclusion. Automation isn't really AI, and at the point the plain text LLMs aren't really "learning" so largely represent a static database. Either way I have yet to ever see an AI output that I would place above the equivalent thing that could be put together by myself using regular tools. And I worry that it's going to make people even more illiterate.

 

[Darqhelmet]

When using it for research you're supposed to validate all of the information with the links it provides you. If it doesn't provide links you need to tell it to do so. If you're using ChatGPT, you can create project folders with a set of instructions that tell it what information and sources you prefer. So you can ask it to pull data from tech sheets, manuals, government websites and the like and it'll provide direct links to them so you can check them yourself. Furthermore, when using it as a research tool you're not supposed to take the output at face value, and you need to continuously push back on its assumptions and have it push back on yours until you create a working model of what you're researching.

Yeah, that is what we did before AI so why not just do that without the extra step that is hurting everyone.

Also is that available for free?

 

[ikeo1]

Eh, I don’t think anyone here is attacking you. Reddit, F-’’em that entire sight is a clown show.

We disagree on AI, not a big deal. Doesn’t mean I think personally think you’re a jerk or evil. Just means we disagree. Everything online is amplified so disagreements seem like a way bigger deal then they are. You can see the person, hear them, etc.

I think I called someone a retard, and got a nono note for it. I text on the screen it’s read worse than it is. in texts i’m a monster I person im easy to get along with.

Not attacking me directly, but their negative animosity towards AI is all over it. I don't take it personally, just interesting to see so many people in different communities with that kind of sentiment. The other dudes i mention, are more annoying and have gone by the wayside now.

 

[Funjeff]

 

[Darqhelmet]

AI makes it so much easier to drone your targets using automatic target recognition. It's supposed to reduce collateral damage but that only applies if you are deploying the correct ordnance. If the operator ends up killing the wrong target(s), he gets to blame it on the AI. Then we also have completely autonomous AI weapons which take the human out of the equation well, except the human targets.

Ya gotta love technology!

Have you seen the little tank drones Ukraine drops behind the Russians line? Things are scary, and supposedly running off AI to choose who and when to fire.

 

[rodralig]

That's not AI it just automated a couple easy Excel formulas that would take the same amount of time to apply yourself if you were familiar. No one is saying that kind of easy task automation is inherently problematic.

Generative AI, as opposed to Agentic AI (which seems what everyone thinks AI is supposed to be), involves performing user-directed analysis on user-provided data, following predefined steps to produce the requested output.

 

[RustVendor]

Generative AI, as opposed to Agentic AI (which seems what everyone thinks AI is supposed to be), involves performing user-directed analysis on user-provided data, following predefined steps to produce the requested output.

Perfect example because you copied a bad definition of Generative AI. By that definition almost any program involving data and an analysis is Generative AI.