I've been reading some Effective Altruist writings and learned that even if AI had moral goals it could be very dangerous .

Like imagine an AI whos goal was to maximize the number of mosquito nets given to Africa, sounds great right?

NO! Because the AI would be willing to do anything in order to achieve it's goals it might do something like commit one of the largest acts of fraud in history to get the money for more mosquito nets!

Hmm... what does this remind me of?

google translate

https://boards.4chan.org/g/thread/99604739

https://github.com/redlib-org/redlib

People are throwing around accusations of compromised github accounts left and right !codecels get in here and start accusing people of being bad actors

UPDATE: THANK YOU TO EVERYONE THAT SUGGESTED USING THE PARALLEL PORT WRAPAROUND CONNECTOR METHOD TO REMOVE THE BIOS STARTUP PASSWORD ON THIS TOSHIBA LAPTOP!

I’ve used paper clips for this and it actually worked! 🖇️🎉 https://t.co/kRWJ4Su49K pic.twitter.com/xcZsPTXJwi

— Bob Pony (@TheBobPony) March 30, 2024

Any tips on getting a job right now?

Could you hire me?

I was thinking about having several bots, but didnt want to hammer the rdrama API so I split the api querying part of my application off.

Its designed to publish events to redis maintaining an nosqlish database of the retrieved items set to time out in 1 hour.

You could subscribe pretty much as many apps as you want to it and that way only one thing is hitting rdrama at a time

Its also leveraging bottleneck, so if you use bottleneck in your API calls back to rdrama and configure it to use the redis instance, you wont exceed rate limiting, and all your applications will share bandwith nicely

I might grab posts or other things, for now Its just focused on comments.

Trump Media shares close down more than 14%, company says Truth Social to launch TV streaming

Earlier Tuesday, Trump Media in a press release said it “has finished the research and development phase of its new live TV streaming platform and will begin scaling up its own content delivery network.”

The company said it will roll out streaming content in three phases, the first of which will introduce Truth Social's content delivery network for streaming live TV to the app for Android, iOS and web.

Phase two will release stand-alone Truth Social streaming apps for phones, tablets and other devices, while phase three will release such apps for home television, Trump Media said.

“The streaming content is expected to focus on live TV including news networks, religious channels, family-friendly content including films and documentaries; and other content that has been cancelled, is at risk of cancellation, or is being suppressed on other platforms and services,” Trump Media said in its release.

I'm sure the company that took 2 years to spin up a Mastodon fork fully will definitely be able to smoothly run a streaming service

• For large screens, centres text so it's not way over on the left (some subs have custom html styles, hence the two sections)
• Expands images automatically
• Hides the image hiphopcirclejerk stuck over their sub (thx j cole drama)

// ==UserScript==
// @name         old_reddit_improver
// @namespace    http://tampermonkey.net/
// @version      2024-01-26
// @description  Make old.reddit.com more usable
// @author       You
// @match        https://old.reddit.com/*
// @icon         https://www.google.com/s2/favicons?sz=64&domain=reddit.com
// @grant        none
// ==/UserScript==

(function() {
    'use strict';
    console.log('---- tampermonkey initiated ----');

    // Remove hhcj pic.
    let earl = document.querySelector('a[href="#EARL"]');
    if (earl) {
        earl.style.display = 'none';
    }

    // Centre page body.
    let body1 = document.querySelector('div[id="siteTable"]');
    if (body1) {
        body1.style = 'width: 70%; margin: auto;';
        console.log('body1:', body1);
    }
    let body2 = document.querySelector('div[class="commentarea"]');
    if (body2) {
        body2.style = 'width: 70%; margin: auto;';
        console.log('body2:', body2);
    }
    console.log('ran centre aligner');

    // Expand images.
    let xpath = "//a[text()='<image>']";
    let iter = document.evaluate(xpath, document, null, XPathResult.ORDERED_NODE_ITERATOR_TYPE, null);
    let array = [];
    let text_elem = iter.iterateNext();
    while (text_elem) {
        array.push(text_elem);
        text_elem = iter.iterateNext();
    }
    array.forEach(i => {
        let img_tag = new Image();
        img_tag.src = i.href;
        img_tag.style.maxWidth = '500px';
        i.parentElement.appendChild(img_tag);
        i.parentElement.removeChild(i);
    })
    console.log('---- tampermonkey complete ----');
})();

• A level 10 OS command injection vulnerability was found in GlobalProtect

• This bug is arguably worse than the Windows-Rust exploit which let attackers execute arbitrary shell commands

Alright so I have a cheap laptop that I loaded debian / plasma onto.

Keyboard cover is wonky and sometimes when I hold left or right arrow it self-inserts a '4', which is fixed with turning off the num pad.

So any time I switch to another window, any time I try to upload something and the popup comes up for me to select a file, the left or up arrow is automatically pressed. Right now it's doing it with the up arrow.

You'd think "you just said your keyboard is messed up why is this shocking" well to that I say here's the weird part.

It doesn't happen randomly, it happens specifically when I change to another window, or a window is popped up like when saving or uploading a file. As soon as I press any arrow key it stops.

For example I'm on my browser now.. if I click to groomercord it will try to edit my last message (up arrow is held), and will keep scrolling up unless I press a random arrow key. When I click back to the browser where I'm writing this post, it begins to scroll up (up arrow held) and again I have to press a random arrow key to stop it.

WTF is going on? Why is me clicking another window triggering an arrow key?!!?! It won't ever happen randomly either. It will never do it again until the window is changed.

To add a weird cherry on top, if I click my terminal (Konsole) it won't scroll up my command history. It won't trigger this weird shit, but as soon as I click to another window it'll trigger the up arrow. First half of today was the left arrow, abot an hr ago it started doing the up arrow.

Pls help solve.

I usually don't game on this laptop, and even when I do, it seems like the games I am interested in are available on there. The only thing I can think of is that I have multiple tabs open in my browser that I plan to one day read. I would lose them if I did it.

I didn't read this. They're Romanian btw.

andres freund: genius programmer responsible for one of the most widely used pieces of software in the world, just single-handedly stopped a huge attack from happening. nyt journalist: 'i am too fucking stupid to understand his work and won't even try but it's very boring!!!' pic.twitter.com/eOw54Kzshb

— anton (𝔴𝔞𝔯𝔱𝔦𝔪𝔢) (@atroyn) April 4, 2024

Self-Taught Optimizer (STOP): Recursively Self-Improving Code Generation

This is the repo for the paper: Self-Taught Optimizer (STOP): Recursively Self-Improving Code Generation

@article{zelikman2023self, title={Self-Taught Optimizer (STOP): Recursively Self-Improving Code Generation}, author={Eric Zelikman, Eliana Lorch, Lester Mackey, Adam Tauman Kalai}, journal={arXiv preprint arXiv:2310.02304}, year={2023} }

Abstract: Several recent advances in AI systems (e.g., Tree-of-Thoughts and Program-Aided Language Models) solve problems by providing a "scaffolding" program that structures multiple calls to language models to generate better outputs. A scaffolding program is written in a programming language such as Python. In this work, we use a language-model-infused scaffolding program to improve itself. We start with a seed "improver" that improves an input program according to a given utility function by querying a language model several times and returning the best solution. We then run this seed improver to improve itself. Across a small set of downstream tasks, the resulting improved improver generates programs with significantly better performance than its seed improver. Afterward, we analyze the variety of self-improvement strategies proposed by the language model, including beam search, genetic algorithms, and simulated annealing. Since the language models themselves are not altered, this is not full recursive self-improvement. Nonetheless, it demonstrates that a modern language model, GPT-4 in our proof-of-concept experiments, is capable of writing code that can call itself to improve itself. We critically consider concerns around the development of self-improving technologies and evaluate the frequency with which the generated code bypasses a sandbox.

https://arstechnica.com/security/2024/03/hackers-can-extract-secret-encryption-keys-from-apples-mac-chips/

However, that data was being shared with data brokers LexisNexis Risk Solutions and Verisk, who'd then sell it to the insurance industry. Earlier this month, the NY Times reported that an owner of a Cadillac XT6 in Florida sued GM after he saw his insurance rates double due to data collected about his driving.