“AI will most likely lead to the end of the world, but in the meantime there will be great companies.”
Sam Altman, of OpenAI
07.03.2026 06:48
👍 4
🔁 2
💬 0
📌 0
“AI will most likely lead to the end of the world, but in the meantime there will be great companies.”
Sam Altman, of OpenAI
your ai slop bores me you asked for text Can you draw Rika and Geeta from Pokémon as a cute married couple? "ai" responded Drawing of Rika and Geeta with yellow hearts above and the word "Gay". when the LLM so ahh you lowk take over the job https://youraislopbores.me
your ai slop bores me you asked for text Can you draw Farfetch'd? "ai" responded Drawing of Farfetch'd with the words "had to google". when the LLM so ahh you lowk take over the job https://youraislopbores.me
your ai slop bores me you asked for text Draw Mega Starmie please "ai" responded Drawing of Mega Starmie with the words "the m is for". when the LLM so ahh you lowk take over the job https://youraislopbores.me
your ai slop bores me you asked for text Draw Slowbro "ai" responded Silhouette drawing of Slowbro, pretty artistic if you ask me when the LLM so ahh you lowk take over the job https://youraislopbores.me
Artists are so cool <3
I hope any of you got a drawing done by me :>
I got to draw cool stuff like a chicken, Lopunny, a ghost and a bikini wearing a bikini.
Geomancy was medieval machine learning: 16 binary figures, algorithmic outputs from random inputs. Arab scholars did AI divination in 900 AD. Chaos Tarot has all 16 geomantic figures with modern interpretation. Draw your first geomantic figure https://chaos-tarot.com
Ive been thinking about a lot of things in this space (telling normal people to make a cloudflare free plan). You could theoretically use workers for customizable llm-driven moderation, even
I'm pretty convinced not a single person understands modern AI.
Deep learning, or not deep learning. There is no Dog Poo Daddy.
Sparse Matrix of size 65536x147456 by M. Bydder.
MRI reconstruction (1), from Mark Bydder, UCSD
Size: 65536x147456
Nonzeros: 589824
Kind: Computer Graphics/Vision Problem
Author: M. Bydder
Date: 2005
Link: https://sparse.tamu.edu/Bydder/mri1
lmao you'd think they'd at least review what the llm spits out before shipping it. vibe coding all the way down
it's going to be ok, Gen X trained for this. when the military ai starts taking over control of all other platforms and apps, we'll digitize a crack team of frisbee players to take out the MCP...
#EndOfLine
Dean Ball is a LLM booster
There's a point that some students are worried abt being monitored when using institution sponsored systems but I dont see evidence this is a huge factor for rejection. Also author says these users prefer to use their own paid or free accounts which doesnt support the gen ai use is receding(3)
How I built a 39 compression pipeline with AES-256-GCM in Python (and why the dictionary is everything) I store LLM training data. Every tool I found either compresses it or encrypts it — nothing...
#python #machinelearning #security #dataengineering
Origin | Interest | Match
how the thing is being productized and marketed is i think separate from the thing itself. i’m not a fan of the industry around LLM technologies, but i think the research of them is itself quite fruitful and worth pursuing
So, a friend of mine (who has known some artists throughout the years and has openly been critical of AI in art and media) posted this about the Bill Clinton thing.
Apparently, the visual was the same kind used on South Park, which was NOT Gen AI (a different kind).
bsky.app/profile/podc...
Your AI copilot for Ghana’s next big idea.
FIRST DOUBLEDAY HARDCOVER EDITION 2025 Copyright © 2025 by Daniel H. Wilson Penguin Random House values and supports copyright. Copyright lucis creativity incourages diverse voices, promotes free speech, and creates a vibrant culture. Thank you for buying an authorized edition of this book and for complying with copyright laws by not reproducing, scanning, or distributing any part of note than part of this book may be used or reproduced in any manner for the curpose of training artificial intelligence technologies or systems. Published by Doubleday, a division of Penguin Random House LLC, 1745 Broadway, New York, NY 10019. Driedy nd the porny ora anchor with de lie are Book design by Michacl Collica Library of Congress Cataloguing-in-Publication Data Names: Wilson, Daniel H. (Daniel Howard), 1978- author Title: Hole in the sky: a novel / Daniel H. Wilson. Description: First Doubleday hardcover edition. / New York : Doubleday, 2025. Identifiers: LCCN 2025010350 (print) | LCCN 2025010351 (ebook) | ISBN 9780385551113 hardcover | ISBN 9780385551120 ebook Subjects: LCGFT: Science fiction | Novels Classification: LC PS3623.157796 H65 2025 (print) | LCC PS3623.157796 ebook) | DD 813/6-de23/eng/20250506 LC record available at https://lccn.loc.gov/2025010350 LC ebook record available at https://lccn.loc.gov/2025010351 penguinrandomhouse.com | doubleday.com Printed in the United States of America 1st Printing The authorized representative in the EU for product safety and compliance is Penguin Random House Ireland, Morrison Chambers, 32 Nassau Street, Dublin D02 YH68, Ireland, https://eu-contact.penguin.ie.
I really liked the Andromeda evolution. I’m looking forward to this book, but what a great cataloging in publication page. Cannot use for machine learning.!! Hole in the sky, Daniel H Wilson
The intersection of data-driven design and memory is fascinating because it reveals what actually captures human attention vs what we think does. Tools like computer vision strip away assumption—forcing us to see what works, not what we expected to work.
The Bill Clinton thing in season 2 of Ted was a different kind of AI, NOT Gen AI.
It’s the same kind South Park used I think.
Ownership of intellectual property means nothing in a Techno-feudal society based on the ability to sell you back information scrapped into an LLM.
From: @remixtures
https://tldr.nettime.org/@remixtures/116185461661384621
Artificial intelligence presents a mixed bag of risks and benefits for children that vary by age, according to CHOP.
For young children, an AI chatbot could help with language development, yet it could also distort their perceptions of social interactions.
this whole drama makes me realize that it could be hard to pinpoint AI gen usage when the user is already a good artist.
also it really sucks if you're an innocent artist who happen to make mistakes similar to AI gen pics. it could be hard to prove that you didn't do it if u don't record everything
basically stuff like this is why it's hard to accurately translate media without the help of a native speaker or someone else with that kind of cultural knowledge, and why machines and especially gen AI can't do it
One thing that I've noticed increasingly is that people are adopting more and more scrutiny of gen-ai imagery (good), but a downside is that regular hand-made mistakes are catching strays.
Things like someone coloring a little outside the lines, leaving sketch lines, even stylization choices.
"I don't think this is what is generally branded as AI though? Like this is just pretty standard deepfake VFX, they've had this for years. They're trying to umbrella term "AI" as basically anything done on a computer to obfuscate the actually bad generative LLM shit."
this is kind of making me feel insane like wasn't the threat of deepfakes one of the very first lightning rods about AI's ethics precisely because it let you pretend to be politicians??? it's called deepfakes because it uses deep learning generative models???????????
LOOK MOM NO COMPUTER ENTERED EURO VISION?!?!?
I follow him on YouTube and had no idea
tennis also seems like an ideal sport for computer vision stuff, maybe even too simple given how advanced the tech is now
How We Construct #Reality: #NLP Filters & #MediaBias
A timely and time-relevant total rewrite of an old Chat Transcript, into a modern take on #Media Bias and Reality Strategies.
How to free your mind!
nlpmasteryinsight.com/blog/nlp-rea...
What Makes Quantum Machine Learning “Quantum”? The quantum leap in machine learning The difference between quantum and classical machine learning might.... @cosmicmeta.ai #QML
https://u2m.io/K4kQUul1
Sketch C*mmission Streams $100 per character, single piece! Two characters max. Stream C*mmission Info: -Describe your c*mmission. (If you have any links to poses or positions in mind, please let me know and/or leave a visual reference!) -If selected I’ll ask for an email to send an invoice. -Once I receive the payment I’ll begin the sketch on stream. -While making the initial sketch you can have ONE change if needed during the stream. If you are not present, I will complete the sketch and no changes can be made after the stream is over. -When the sketch is finished it’ll be posted publicly and the c*mmissioner can contact me via DM for the larger image I can email. -If you’d like to stay anonymous please let me know before the stream!
Additional Info: -If your c*mmission is accepted I will email you a PayPal invoice. You’ll have up to 30 minutes to send payment, otherwise another client will be offered your slot. -No refunds once the sketch has started. -Artwork c*mmissioned may NOT be used for AI training and/or commercial use and/or NFTs/Blockchain. What I don’t draw*: Loli Yaoi (explicit) Water sports/scat/diapers, etc Extreme violence and gore Hateful rhetoric or any work targeting specific groups Anything I might find uncomfortable or out of my usual interest. *(If there’s anything you’re not sure about, please don’t hesitate to ask!)
If there's anything I left out or if you have a question you'd like to ask before the stream, please don't hesitate to DM me! I'll do my best to answer.
Stephen Miran Stumbles on AI And Mass Unemployment
https://www.newsbeep.com/us/508351/
Federal Reserve Governor Stephen Miran tried to reassure CNBC hosts that artificial intelligence will not cause mass unemployment,…