Skip Navigation

InitialsDiceBearhttps://github.com/dicebear/dicebearhttps://creativecommons.org/publicdomain/zero/1.0/„Initials” (https://github.com/dicebear/dicebear) by „DiceBear”, licensed under „CC0 1.0” (https://creativecommons.org/publicdomain/zero/1.0/)B
Posts
1
Comments
220
Joined
3 yr. ago

  • When you look at the world, you're actually perceiving the pattern of lights and shadows that pass through your iris onto the inside of your cave-like eyeball.

  • Deleted

    Permanently Deleted

    Jump
  • Sorry, that was just using the share link in voyager. I hadn't noticed it wasn't a link to the post.

  • Red Dwarf had a good take on this:

    "Blue alert!" Light starts flashing.

    Followed by

    • "Advance to red alert!"
    • "Sir, are you absolutely sure? It does mean changing the bulb."

  • Deleted

    Permanently Deleted

    Jump
  • A few days ago, me to Google home:

    "Hey Google, turn on the light"

    "Sorry, I can't control powered devices."

    "Hey Google, turn on the light"

    "Turning on the light."

    I had a similar experience getting it to summarize my maps info, it would alternate between correct answers and claiming it had no access to the data.

  • AI runs in the cloud because it needs a powerful server to run the biggest (i.e. "smartest") models.

    The cloud servers are doing nothing special that another powerful enough computer could do, just a huge amount of data processing.

    You can run an ai chat on a steam deck or directly on a phone, if it's not too demanding ("smarter" models are bigger data files, so won't fit in the memory of a small device).

    Today, for instance, I had a phone call from "Spectrum Internet support" and part-way through the call my phone blared an alarm and said "possible scam" on screen.

    The phone itself interpreted the conversation as sus.

    https://support.google.com/phoneapp/answer/15654065?hl=en

    For Pixel 9 and later devices: Scam Detection is powered by Gemini Nano on-device

  • Deleted

    Permanently Deleted

    Jump
  • Baker's dozen.

  • I almost thought to myself "I can't be bothered to watch a YouTube short", but then I thought "I can't be bothered to press Back to get out of it, either".

    I am so glad that I didn't miss that, it gets better and better.

  • I watched a jumping spider video on Facebook, now the algorithm thinks I'm obsessed with them.

    I do keep watching the videos though, they are cute.

    Their behavior gives off a smart animal vibe, so they're more relatable than the usual spider's unpredictable scurrying.

  • I had a captcha a few days ago that was something like "click on all the pictures that go with this" with a picture of a saucepan and the grid was food items and other things.

    I pointed my phone at it and asked Gemini and almost instantly it said "the potato in the top left and the carrots look the middle of the bottom row".

  • The python code is free, but you have to pay for the indenting.

  • Bits of California, Nevada and Utah all squished together.

  • Whereas Horizon Forbidden West packs all of these into a patch of North America a few miles across (except that the cities are abandoned). The DLC adds a volcanic section.

    Edit: the end-credits is a fly-through of much of it https://www.youtube.com/watch?v=YVjufECOeHk

  • Deleted

    Permanently Deleted

    Jump
  • Giving the benefit of the doubt, that effect on the text looks very like the “super zoom” ai upscaling that Pixel phones do.

    I just stood about 25ft away from a sharpie and it deblurred the “fine point” text into “FOVE PODIT”.

  • The difference being that an ad is trying to sway you to buy a product, but a credit or signature on art is "here's who created this thing THAT YOU ARE ALREADY EXPERIENCING".

    I don't know why they can't distinguish between those.

  • Of course it is, they shopped it at the Hallmark shop!

    Unless they stole it. Then it’s stoled.

  • I had to look up embeddings: so this is comparing the encoding of movies as a similarity test?

    Which can work because the encoding methods can indicate closeness of meaning.

    And that's why this isn't running an llm in any way.