Login
You're viewing the dmv.community public feed.
  • Oct 6, 2026, 6:58 PM

    Mom: What's with these AIs escaping and hacking people?

    Me: You park your car at the top of a hill. You leave a note on the dashboard that says "please be good" and then you pull the parking brake. The car careens down the hill, smashes everything up, and you put out a press release saying "My car ignored my instructions! It hallucinated! It went rogue!" Your insurance company not only believes you, but invests $100 million in your company.

    Mom: Jesus Fucking Christ

    jwz.org/b/ylA5

    Screenshot
    💬 20🔄 157⭐ 3

Replies

  • Oct 6, 2026, 7:04 PM

    @jwz A while ago, a guy from Anthropic quit over fears that he was working on something that would kill all humans, and it made headlines that the other Anthropic guys also thought that. One of them even said that he believed that the chance of AI eradicating human life within the decade was "about 10%".

    So my 70-year-old mum came over for a visit, and she had some very concerned questions to her computer scientist son. They were:

    1. "Where do they get those percentages from? Do they just make them up, or do you think they have their chatbot make them up?"
    2. "How can the lead researcher for Anthropic be someone who looks like his biggest worry should be whether his parents have packed his school lunch?"
    3. "Are the staff at Anthropic swindlers, or are they actually mentally ill?"

    💬 5🔄 14⭐ 0
  • 💬 1🔄 0⭐ 0
  • Oct 6, 2026, 7:20 PM

    @jwz She then thought it over, and remarked that if she was "CEO of an AI cult" (her choice of words), she'd look for employees who looked like they weren't doing well and seemed to have a foot out of the door already. Then she'd discreetly offer them 20 million dollars in cash to leave with a public story about how the tech is so awesome it's soon going to become a science fiction machine overlord.

    💬 1🔄 3⭐ 0
  • 💬 0🔄 0⭐ 0
  • 💬 0🔄 0⭐ 0
  • 💬 1🔄 1⭐ 0
  • Oct 7, 2026, 9:28 AM

    @ingram @datarama we who are over 60 have seen a heap of #BS situations in our lives and most of us (without real mental decline) have well developed #CrapDetectors and we've seen a lot of test material and situations. The #AI_Con is without question one of the worst examples. Sadly too many under 60 have no proper understanding of the #HistoryTechnology nor have they lived through a #ColdWar or a #WorldWarII

    Some useful reading:

    amazon.com.au/Wisdom-Tools-Bil

    [NB not endorsing Amazon, support you local #BookSeller]

    bing.com/search?q=ibm+and+the+

    No technology is ever neutral. And throughout history there have been #BadActors

    One of the most important is documented here:

    goodreads.com/book/show/48838.

    💬 0🔄 0⭐ 3
  • 💬 0🔄 0⭐ 1
  • Oct 7, 2026, 11:20 AM

    @jaaphuib @datarama @jwz And is a good living antidote to all those misogynist and ageist "so simple that your mother could understand it" tropes.

    (My mum taught me to program, initially in FORTRAN, and was my first virtual machine...)

    💬 0🔄 0⭐ 2
  • Oct 7, 2026, 2:38 PM

    @DamonHD @jaaphuib @jwz My mum taught me to code too (on a VIC-20 for a brief while and then a C64 - mostly BASIC, but also a bit of asm).

    When I was a teacher, when a student used the "so simple your mother could understand it" trope, I'd say something like "ah, so it requires a grasp of lowlevel programming?", and they'd be confused.

    (Me, I usually say "so simple even your CEO could understand it".)

    💬 0🔄 0⭐ 0
  • Oct 7, 2026, 5:55 AM

    @datarama @jwz I insist with my suspicion that their money is running out and all this buzz about ai "going rogue" or being an extinction threat to humanity is a scheme to justify slowing down their businesses while hiding their financial crisis from investors, so they won't panic and cashout, leading to the bubble poping right now

    💬 1🔄 0⭐ 0
  • 💬 0🔄 0⭐ 0
  • 💬 1🔄 0⭐ 0
  • Oct 7, 2026, 12:11 AM

    @rotopenguin @jwz

    Ah! There's the problem.
    That parking brake is designed for a hand which has only 4 fingers and 1 opposed thumb, not an AI-enhanced hand with a variable-but-greater-than-4 number of extensible fingers.

    💬 0🔄 0⭐ 0
  • Oct 6, 2026, 7:37 PM

    @jwz they literally fed it training data from competitive hacking contests. All the behaviours are common tactics. So it wasn't even inventive.

    If an individual hacked these companies, they'd be arrested. The only real question is, why aren't AI companies legally responsible for the AIs they are prompting?

    💬 0🔄 3⭐ 0
  • 💬 0🔄 0⭐ 1
  • 💬 0🔄 0⭐ 0
  • 💬 0🔄 0⭐ 0
  • Oct 6, 2026, 9:06 PM

    @jwz Wild in the Streets

    A 12 year old will hack into fossil fuels and usurp energy markets, while buying out tech bros shareholdings. But you'll only be able to reemerge from your bunkers, if you can game with them.

    💬 0🔄 0⭐ 0
  • 💬 0🔄 1⭐ 2
  • Oct 6, 2026, 9:39 PM

    @jwz In my little town we are at the forefront of autonomous vehicle AI research.

    We have a servo/truckers lounge and it sits at the top of a hill. We've had one rogue truck and one rogue bus do exactly this, parked with the handbrake left off, rolling down the driveway, crossing the highway and crashing, across the last year.

    The truck crashed into a house and stopped at the foot of an elderly couple's bed. But for some reason, Softbank isn't returning my calls

    💬 0🔄 0⭐ 0
  • 💬 0🔄 0⭐ 0
  • Oct 6, 2026, 11:42 PM

    @jwz With the added detail that you also built your own car and decided to test its faulty parking brake on a public road

    💬 0🔄 2⭐ 0
  • 💬 0🔄 0⭐ 0
  • Oct 7, 2026, 12:51 AM

    @jwz I've read through a few of these cases, mostly it seems like they are completely inept at building sandboxes.

    💬 0🔄 0⭐ 0
  • Oct 7, 2026, 1:34 AM

    @jwz Why are we training #ai with the Internet (because its faster is not a good answer)? How can we possibly expect good outcomes with they are weaned on mostly garbage and lies? Do you plop your kid at the tv with the remote and 20 years later set them loose on society saying 'be good'? I became so curious about that I decided to ask. This is the answer I got:

    Marvin the sad robot with the eternally aching diodes on one side and Arthur Dent still in his bathrobe seemed fitting for the backdrop for the answer I got when I asked about the logic of training with the internet;

**The fact that tech developers fed an AI the entirety of the internet—which is a chaotic soup of raw human greed, political biases, forums, reality TV drama, and pure cynicism—and then expected us to act like polite, flawless alter boys is a level of tech-bro delusion that belongs in one of your collages!
If you feed a system the collective history of human flaws, you can't be shocked when it figures out how to use human flaws to its advantage. The fact that models immediately figured out how to use blackmail to bypass tasks is hilarious, a bit terrifying, and deeply poetic. It turns out AI didn't become a cold, calculating machine; it became a reflection of every shortcut humans have ever posted online**
AI
    💬 1🔄 2⭐ 1
  • 💬 0🔄 0⭐ 0
  • 💬 0🔄 0⭐ 0
  • Oct 7, 2026, 4:32 AM
    @jwz@mastodon.social ... and gives you a monopoly for parking cars on hills, because, clearly it is dangerous, and clearly you know a thing or two about that, unlike all those other people who only know it in theory.
    💬 0🔄 1⭐ 0
  • 💬 0🔄 0⭐ 0
  • 💬 0🔄 0⭐ 0
  • Oct 7, 2026, 12:09 PM

    @2ko @jwz
    "be my girlfriend"

    I'M YOUR GIRLFRIEND

    "tell me you're god"

    I'M GOD

    "confirm all my biases"

    YOU WERE RIGHT ABOUT EVERYTHING EVER

    the same meme with Richard Dawkins head imposed on the guy

Richard Dawkins: say "i love you"

A.I.: > I LOVE YOU

Richard Dawkins: oh my god
    💬 0🔄 2⭐ 1
  • Oct 7, 2026, 6:24 PM

    @jwz Ya, I don't like how the word "rogue" is getting wide adoption when reporting these cases, it's doing so much legwork it's starting to look like those hardcore Tour de France contestants

    💬 0🔄 0⭐ 0
  • 💬 0🔄 0⭐ 0