My FB is started to get these creepy AI clips of actors from old TV shows then and now with the old character and what they look like today in live action. If the actor is deceased it shows then the "today" image has them wearing angel wings.



Then there is crap like this ad for stupid maps.

Modern statues on campus

The samey-ness of AI scripts.

Claude! said:

---

The samey-ness of AI scripts.

---

Well, AI is at its core a statistical engine and it incorporates a while craoload of very average writing by very average people in its source data. It's almost inherently mediocre at art and composition because it's averaging in a whole hell of a lot of average to generate output. And unfortunately, they also use a lot of incorrect or bad data on top of that.

I'm not trying to disagree with you or provide counterpoints. I am genuinely curious what your thoughts are and feel like you have good reason for them.

What are your thoughts on someone using AI to organize content but not having AI write it from scratch? So, the person gives AI a wall of unstructured notes, reasoning, logic, outcomes, etc. on a particular topic and then intentionally prompts it to organize all of that into a certain structure or layout. Training materials or user guides would be an example.

Full candor, I do this a lot as organizing and structuring things is often the biggest struggle I have when creating documentation, which I do a lot. I have all the ideas and knowledge in my head, but it feels tedious to organize them in a clean way. I have been a bit self conscious about how much I use AI for the final product, but the efficiency tradeoff has won out.

I dont think there is anything wrong with that at all.. For instance, asking AI to provide the outline, but then you filling out the meat? Totally fine.

Having an entire email filtered into AI to sound like a generic pep talk written by a robot? Nah. When its obvious that a person just wrote a pompt, then copied and pasted the result, it feels like they don't respect your time or trust their own intellect.

we don't need this to be our lives in the workplace:
https://www.instagram.com/reel/DRoSCD5kbYH/?hl=en

its the super hokey canned stuff that makes me insane.. or analysis that if I wanted chatgpt to do, I would have asked chatgpt, not my colleague.

I keep wanting to provide specific examples of lines in emails that have been sent, but I also don't want to be so specific that someone might connect some dots.

AI garbage just can't quite match Shirley Manson's voice.

No logical person likes AI. At least not in this current form.

This guy has a great channel.

My biggest complaint is having sift through MegaLops posts claiming everything is "AI slop".

CharleyKerfeld said:

---

i had a potential client tell me that he was convinced he had an idea for the greatest song ever written that was guaranteed to make him a millionaire because ChatGTP had told him so and it was a "conglomeration of all the smartest people in the world."

I sent him a transcript from a prompt of me asking ChatGTP if I could make $1 million selling bananas out of my living room and it wholeheartedly agreed that it was a great idea.

---

Ask ChatGTP which idea would be the most appeeling.

Z3phyr said:

---

Nothing "bothers" me in the emotional sense but there are a few things about AI that are objectively… messy, and worth keeping an eye on:

1. Confident wrong answers (the big one)
AI can sound very sure while being completely off. That's dangerous in fields like engineering , medicine, or law because it feels trustworthy even when it shouldn't be.

2. Garbage in garbage out
AI reflects its training data. If the data is biased, outdated, or just wrong, the output inherits that. It doesn't "know better" it averages patterns.

3. Over-reliance
People start skipping first-principles thinking. Instead of checking calcs or code requirements, they trust the output. That's how you end up with a system that "looks right" but fails in the field.

---

It feels like a precocious new college grad to me. Usually right, very capable, always certain. Good with handling non-critical tasks, needs their work checked especially for anything important. Good for idea generation and new angles. Definitely shouldn't be blindly trusted.

I started using it (Google Gemini) for diet and exercise tracking and advice. It's mostly good. Issues I've had with it: struggles knowing what time of day it is, I have to watch it on calorie tracking because it has a bias towards telling me I'm meeting my goals when I'm really over or under, and it can be long winded and repetitive. Working on training/programming those errors out of it. It's particularly good at coming up with meal planning ideas and providing positive feedback. It's kind of strange to me that a computer is better at the creative stuff than it is at the computing stuff.

Nvm.

In the funnier version of this, the last frame is pulled back and you see Padme with giant fake AI cleavage.

Call me unoriginal. IDGAF.

I use it every day to summarize lease agreements or break down a contract. I also use it to formulate emails to large groups because why should I devote my valuable time trying to impress some old hag on my creative writing abilities?