You must log in or # to comment.
If this truly needs to be said, a giant meteor is too good for us. I get that people are stupid as fuck but Jesus, trusting your life to AI at this point is next level idiocy
Wtf
I get that it’s a bad idea to trust AI to something so important. And that it’s probably going to shock or kill you.
But since I’m not an electrician, ELI5. What’s wrong with this diagram?
This diagram is the same as sticking a fork in all 3 holes on an outlet.
In addition to what the other person said, there’s a wire here connecting the two prongs together. This is a short circuit and would at least blow a breaker and at worst start a fire.
The outlet has three wires (black, white, and green), and the device to be plugged in has three wires (black, white, and green). The whole point of the plug is to connect each color wire to its counterpart, and only to its counterpart.
This picture is so frustrating. There’s no apparent source for the “live” wire. But if you tried following this with your real wiring (white, black and green do exist), you would be connecting your hot, neutral and ground lines all together. I doubt it would kill you at 120 V but it would definitely trip your breaker the moment you plug it in, and almost certainly make some nice big arcs.
Its ok, the AI predicted this and oriented the prongs incorrectly so there is no chance to accidentally plug it into an outlet.
I showed this to my electrician dad. He had a chuckle.
You know this makes you all look like idiots right?
It’s watching a bunch of morons trying to hammer in a screw and saying “look it doesn’t work”.
Darwin awards?
DO NOT TRUST ANY AI TECHNICAL DRAWING!
It was legit a little scary how long I stared at this diagram in utter bewilderment.
It’s 2025 and the tech bros have made it where I can’t believe my own lying eyes.
DO NOT TRUST ANY AI
TECHNICAL DRAWINGFTFY
“Never trust a computer you can’t throw out a window” - Steve Wozniak
There been so much gaslighting I keep thinking I’m missing something as I keep doing the mandatory AI usage from work and largely throwing away the results. Yet I keep repeatedly hear people hush about how great it is.
Last week I finally had something that I thought would be right up it’s alley: a fairly milquetoast program that needed to be ported from one language to another. It generated the ugliest unmaintainable code I could imagine. I went to execute the test suite and it failed every single one. The code was ugly, hard to follow, failed to important a number of sanity checks, kept repeating common code without factoring, making up calls that didn’t exist. In trying to damage it I decided to ask it to implement a super simple check that should have been a one liner and it suggested splatting over a hundred lines of code all over the file to randomly mess with imaginary end points for no apparent reason and still falling to just check the length of the array I told it to.
Meanwhile people at my company who don’t even commit any code are writing on internal blogs about how transformative it is, how much it makes their coding better (that they didn’t even do) and executives are commenting their vision and intelligence and promoting them…
I actually tried to get an AI to code a simulation for AI agents. God, it’s so shitty at everything it does, it is only good to torture for entertainment…and to teach Excel apparently, only uses I found for it.
I will say there’s a few common boilerplate things that is easier to express as quick prose and get 4-5 lines out of it, highly predictable lines that I could have probably done myself, a little more convenient when I’m rusty on a language or library. Also code completion, same thing of only good for like 3-5 lines at a time.
Still have to watch it like crazy and carefully fix the mistakes and still throw out what it suggested and it can be pretty annoying, but I’m willing to believe it saved me a smidge of time on those. But anything beyond that and it’s a horrible mess I would never abide that’s more trouble to try to fix than just scrap and start over. It’s hard to stay vigilant and micromanage the details as I’m used to reviewing human code and the sorts of mistakes the LLMs make are distinct from what humans make.
There are a few things it does well, like movie suggestions based on a list of shows you like. And its good at reformatting data you feed it. But It’s crap overall. Often takes more time to correct it, than starting from scratch.
Definitely feel the waste of time trying to salvage junk it generated. At best I might have it take a horrible stab at it and then just look for key bits to Internet search, read documentation, and do it the actual right and maintainable way.
3-5 lines of code or has a decent chance at being usable or needing a minor tweak. More than that and it just sucks.
Also did use it with some messy list to find essentially duplicate things that weren’t exactly duplicate. It wasn’t 100%, with false negatives and positives, but it let me find a lot of them easily and the situation didn’t require me to find every single duplicate.
JUST. WHY. There are so so many freely available photos already online that don’t need to be generated AGAIN
But those images are hosted by companies other than Google. You might visit those sites instead of staying on a Google-owned domain where they can keep making money off of you.
With deeper integration of AI into search machines, such as Google, the distinction between AI generated and real search results becomes more and more complicated.
Instructions unclear, balls are tight and I can feel a 60Hz hum. When do I insert the neutral wire?
Bend over and I’ll show you
Why is it on the floor next to the soap I just dropped?
Because safety standards in porn are for rubbery devices only. Metal ones are juuust fine, now stop fluffing, our star needs to get back in there!
Unrelated question, why is this post a full fucking screenshot of some dude’s phone uncropped?
Cropping is a lost art that was under appreciated in its time. Oddly, I see it done less even though it’s become substantially easier to do over time, with the tools often built into whatever took the screenshot.
You guys are trusting AI?
Nope. Actual sentient beings aren’t even usually trustworthy so why would I trust a word guesser?
I mean what the fuck am I even seeing?
and also, spicy ground.
Honestly doesn’t look particularly dangerous. I think this nets you a ground neutral connection with two test points?
Would it be useful? Absolutely not. Dangerous? p Probably not.
For reference / non electrical engineers / visual learners:
https://engineerfix.com/electrical/plugs/how-to-wire-a-us-plug/
NA 2 Prong, Type A
NA 3 Prong, Type B
‘Earth’ = Ground = the 3rd, round prong
You can maybe see the AI has kinda sorta merged elements from both of these.
The main problem is that uh… notice how in these real diagrams none of the lines are looped into each other, they are 2 or 3 distinct lines that feed into whatever is being plugged in…
… whereas in the AI diagram… it just forms a continuous loop.
Which… basically means that what the AI labels as ‘neutral’ and ‘ground’… aren’t.
Doing that would instantly cause a short cicruit and potentially immediately start a or multiple fires, trip circuit breakers, etc, depending on the voltage/amperage avalaible to the outlet this AI ‘death-plug’ ie connected to, and how long it takes the circuit breaker to trip.
… thats my layman understanding, corrections from any actual EEs would be appreciated if I’ve mucked something up.
Just a heads up, those diagrams are also insufficient on their own. They don’t indicate which line goes to the wide blade and which goes to the narrow blade on the plug. Getting that backwards on some devices can cause switching and fusing to be on the wrong line and defeat certain circuit and safety protections.
Much as I encourage people to fix their own things, wiring AC mains connections is one of those “if you have to ask, you probably shouldn’t” kind of things.
Thank you, those are very good points to add!
I managed to be busy nearly all day, glad other people have upvoted your addendum.
“You will die”.
Sounds like AI is bringing back natural selection.
Is it me, or do people actively celebrate the death of weak people? It is quite unnerving.
Actively celebrate?
No.
Feel zero sympathy?
Yes.
anyone who asks AI for any help at all.
