Skip Navigation

User banner

jarfil

@ jarfil @beehaw.org

Posts
5
Comments
734
Joined
3 yr. ago

Programmer and sysadmin (DevOps?), wannabe polymath in tech, science and the mind. Neurodivergent, disabled, burned out, and close to throwing in the towel, but still liking ponies 🦄 and sometimes willing to discuss stuff.

  • What the fuck is happening at Reddit?

    Jump

  • Reddit's moderation bots have been extremely trigger happy for many years.

    I got my main account, 10+ years club, suspended... appealed it, and got banned. Then every account I had ever logged into with the same IP, app, or browser as the banned one, at any moment in the past, got banned in cascade.

    Once you get on Reddit's bad side, there's no going back. Suspensions add flags to Reddit's internal "shadow profile" of every account ever linked in any way. They all become more likely to get flagged and suspended, which gets them flagged even more in turn, until Reddit's ban-evasion system kicks in. Then, they're all toast.

    To add insult to injury... once triggered, the bots go back checking your history, applying the most recent moderation guidelines retroactively. Over the following months, the account kept getting notifications about old comments being removed, followed by subreddit bans.

  • Does AI need to be perfect to replace jobs?

    Jump

  • There's a good commentary about that in here:

    AWS CEO Matt Garman just said what everyone is thinking about AI replacing software developers

    “That’s like, one of the dumbest things I’ve ever heard,” he said. “They’re probably the least expensive employees you have, they’re the most leaned into your AI tools.”

    “How’s that going to work when ten years in the future you have no one that has learned anything,”

    https://www.itpro.com/software/development/aws-ceo-matt-garman-just-said-what-everyone-is-thinking-about-ai-replacing-software-developers

  • Authors celebrate “historic” settlement coming soon in Anthropic class action

    Jump

  • Here we go, more Mickey Mouse fueled BS. Instead of fixing the preposterous "until author's death + 70 years" copyright term, the result is a world where tearing up books to train AI is legal, and a class lawsuit settlement with "7 million claimaints" who will get none of it.

    Lawyer circus, is what this is.

  • Stop children using VPNs to watch porn, ministers told

    Jump

  • Stop children using VPNs to watch porn, ministers told

    Jump

  • Probably more helpful to say "Stop using VPNs to watch porn"... helpful for VPN providers' sales, I mean.

  • After using ChatGPT, man swaps his salt for sodium bromide—and suffers psychosis

    Jump

  • Education is supposed to teach "how to learn to learn".

    Left to his own devices, then, without knowing quite what to ask or how to interpret the responses, the man in this case study "did his own research"

    The whole thing with "do your own research", is kind of funny:

    • some use it to avoid explaining their points
    • others use it to come up with a lot of nonsense
    • while the proper way to begin any "research", is to... ask an expert.

    Nobody has ended up in a psych hold, just by reading a bunch of Wikipedia articles, asking ChatGPT... then consulting a doctor.

  • How many r are there in strawberry?

    Jump

  • Not sure if I'm not explaining myself, or you're choosing to not understand me. I'm going to leave it here.

  • How many r are there in strawberry?

    Jump

  • Kind of like saying that ChatGPT is people adding an AI player to the deterministic program of a chat... nah, I'm not going to discuss that. Tic-tac-toe is a classical example problem for neural networks 101, kind of a "hello world".

  • After using ChatGPT, man swaps his salt for sodium bromide—and suffers psychosis

    Jump

  • Good news: advances in medicine have reduced "physical" natural selection so much, that "intellectual" natural selection is overtaking it.

    Now, if only all countries could say the same.

  • AI Is A Money Trap

    Jump

  • Keywords: NPU, unified RAM

    Apple is doing it, AMD is doing it, phones are doing it.

    GPUs with dedicated VRAM are an inefficient way of doing inference. They've been great for research purposes, into what type of NPU may be the best one, but that's been answered already for LLMs. Current step is, achieving mass production.

    5 years sounds realistic, unless WW3.

  • Wikipedia loses challenge against Online Safety Act verification rules

    Jump

  • If the current rate of execution of Project 2025 is an indicator... no, there are no "decades" left.

  • LLMs’ “simulated reasoning” abilities are a “brittle mirage,” researchers find

    Jump

  • chain-of-thought models

    There are no "CoT LLMs", a CoT means externally iterating an LLM. The strength of CoT, resides in its ability to pull up external resources at each iteration, not in dogfooding the LLM its own outputs.

    "Researchers" didn't "find out" this now, it was known from day one.

    As for who needs to hear it... well, apparently people unable to tell apart an LLM from an AI.

  • How many r are there in strawberry?

    Jump

  • Trump Is Launching an AI Search Engine Powered by Perplexity

    Jump

  • Looks fine so far.

  • Trump Is Launching an AI Search Engine Powered by Perplexity

    Jump

  • Truth

  • Google’s healthcare AI made up a body part — what happens when doctors don’t notice?

    Jump

  • [sarcasm] Too many words bad... next time, try this one: "Whaaat?" [/sarcasm]

    Seriously, people really need to learn to use AI as a tool, not as an omniscient oracle, and not as an idiot baby.

  • you got it, boss

    Jump

  • Kind of did that already, and for less money.

    On the bright side, daggers are cheap... so if you really hated them that much... prison is like a vacation where you get to read a lot of books, right...?

  • How many r are there in strawberry?

    Jump

  • How many r are there in strawberry?

    Jump

  • Wikipedia Editors Adopt ‘Speedy Deletion’ Policy for AI Slop Articles | 404 Media

    Jump

  • Of course. I also hope this will stop like 99% of the skiddie spam. I'm just afraid that, like it has happened with hacking in general, a noob installing Kali will get a ton of one-click ways to bypass these measures... and then, what's next?

    Genai inserting watermarking would be great, but that's hard to do with text, in any way that isn't easily removed.