Privacy-First: No cameras required - uses WiFi signals for pose detection
That's not how privacy works.
Privacy-First: No cameras required - uses WiFi signals for pose detection
That's not how privacy works.
I have a visceral "AI" sensor that triggers when I see these:
"Rust Implementation (v2)"
"Performance Benchmarks (Validated)"
Human beings don't self-validate explicitly like that. AI loves doing it.
You generate code, there's a bug, you ask for a fix, your AI of choice will always output with:
*** Fix build issue ***
*** End fix ***
and then call it "Version 2 (Validated)".
Sometimes it's more subtle, but you can feel it, it loves adding "confirmed", "working", "validated".
This comment has been confirmed and validated by an actual human being 👍
My sensor is much simpler. If I see emoji in headings or bulleted lists, I assume it's shit. It might be AI slop, or it might just be kids getting overexcited with the little pictures, but both deserve suspicion and scrutiny.
If a bunch of the emoji don't even make sense it can get in the bin.
This comment is so true 🚀🚀🚀
Ahhh idk, I saw a lot of genuine repos do emojis, at least for headings. Even before LLMs.
I like them 'cause with the right amount, it makes a README easier to parse when quickly scrolling over it.
My changelog generation tools output emojis because our lives are too short to not use 🚀
As an ancient husk of a person, it all looks crack-addled to me. I don't really see how you can parse out headings from emoji because their usage isn't consistent.
I like putting the little pictures in my readmes sometimes. In my biologically generated repositories. Please don't discriminate against neat little pictures you can just put in text 🐑.
I have a project with a bunch of compose files that define the services I self host. I "deploy" the project by sshing into my server and doing "git pull" which means I'm often making changes that don't get tested before committing to source control. As a result I have long chains of commits like:
And now I'm wondering if I've been an llm this whole time
No the AI would have called it fixed, “production-ready,” committed, and pushed after the first refactor.
Make your changes in a new branch and rebase/squash when you push it to main.
This also means modifying your git pull command to pull the correct branch. A small change perhaps, but may be harder than just committing to main lol.
I had a similar problem with GitHub actions, it was hard to test without messing up the main repo history.
Why not just edit the YAML directly on the server via a command-line text editor or SSHFS and then push from there when it works?
👉: mission acquired
👊: bugs squashed
👍: code validated
👏: congratulations on this exquisite piece of software
✍️: ready to do more!
I am no programmer and understand almost nothing of the documentation and yet somehow I can tell it's all bullshit.
It reads like a kid making up words in an attempt to sound smart mixed up with the description for a shady Amazon product.
I guess it's reading comprehension. Utter bullshit reeks the same regardless of the field.
That's absolutely awesome!
I'm gonna start referring to this as 'smelling AI slop'
You got the sense to sniff it out, even without programming experience. And that's a damn good sense to have these days 👍
Throughput metrics
Phase Sanitization 67-85 Melem/s
😆
(Turns out it does exist! But it's just a chemical https://en.wikipedia.org/wiki/Melem )
Forgot to put "make sure the project compiles" in his .md files. What an amateur.
I mean well before AI, it was pretty common that a GitHub repo wouldn't compile.
Performance Benchmarks (Validated) yup, 100% totally validated. It's like when you buy something thats wayy too cheap for what it should be off of Temu and it shows up with a QC and Validation card that they clearly just print on a large sheet and cut down that says QC OK
Everyone’s talking about the different things that give it away and here I am with “WiFi dense human pose…” wtf
You can track/identify people in range of a wifi router based on how the wifi signal is disrupted.
I believe that the original people claimed you could ID individual people using their approach, but I suspect that's under ideal conditions and/or with some training against individual people.
anyone gonna cop the $1500 hour session for agentic engineering

And this is why I come up with stupid concepts like this, coding without numbers, just to fuck up artificial intelligence...
https://lemmy.world/post/43158470
Sorry I said it was Friday, it was actually Sunday 🤦♂️
To be honest, I don't think that changes anything. Once you got the relation Z = 0, O = 1, etc., or whatever symbol you represent "1", "2", etc., you just have to do algebra.
Also, how is AI gonna interpret such code without running it first?
Its kinda easy for human eyes to see what I did, but how is AI gonna comprehend what letter equals what number?
It doesn't comprehend anything. That's the damn point.
Though it WILL "understand" what you did by the algorithms that break down code turning the variable into another token. So all you're really doing is costing yourself more time and money in the slop machine.
Exactly. But it totally throws off AI..
I've got a more extended version I'm still casually working on when I get bored, which includes ELRUIYDAMB
Care to guess what those letters mean numerically?
I'll drop a clue, B=Billion
"In the fast-paced world of artificial intelligence, I'm constantly finding new ways to harness the power of technology. One of the most captivating aspects of AI models like GPT is their ability to "hallucinate" – generating completely new ideas and concepts that go beyond mere data processing. This capability underscores AI's potential to create, not just analyze."
First time I am seeing someone sell hallucination as a "feature not a bug"
~24000 stars as of now. Stars might possibly be bots too or techbros on AI overdrive
GitHub could ban a lot of bots of they took a close look at that repo and who starred it. Bet they will. Mmhmm. Yup. Any time now.
Based on the description, I'm glad it doesn't work....
"Production-ready"
I totally believe it's ready for production at Microslop Corporation. They need a larger % of code written by AI, of course.
Acknowledgments: Based on some research somehow maybe by some guy I guess…
I didn't really need another reason to keep my AP firmly disabled, but thanks I guess.
How many routers are required for this? I see it mentions that they should overlap, and there's a diagram that seems to imply 3 routers, but I'm not seeing it saying what the minimum is or if more makes a difference.
top 62 comments