Spyke

Replies

Comment on

Unpopular opinion

Reply in thread

Left side: Black mirror S01E02 "fifteen million merits" . A guy tries to "break the system" but this backfires and his critic that was supposed to change people's minds is absorbed by it and turned into an entertainment product. The upper-left image shows the moment in the episode where he criticizes the system threatening to kill himself while the bottom one shows the final image of the episode where he how lives in an expensive suite

Right side: "Being ugly : My Experience" A youtube video of a guy explaining how his unattractiveness has biased his life and brought unhappiness upon him. A reason why this became viral, besides the obvious connection by many due to the topic was that a girl commented that she found the guy of the video very cute and they actually became a couple

The meme: It compares both cases implying that the guy on the right was breaking the system but that his cause was "silenced" by providing him a girlfriend and turning him into a channel that lectures people on having hope about the prospect of finding a suitable partner

Comment on

This new data poisoning tool lets artists fight back against generative AI

Here's the paper: https://arxiv.org/pdf/2302.04222.pdf

I find it very interesting that someone went in this direction to try to find a way to mitigate plagiarism. This is very akin to adversarial attacks in neural networks (you can read more in this short review https://arxiv.org/pdf/2303.06032.pdf)

I saw some comments saying that you could just build an AI that detects poisoned images, but that wouldn't be feasible with a simple NN classifier or feature-based approaches. This technique changes the artist style itself to something the AI would see differently in the latent space, yet, visually perceived as the same image. So if you're changing to a different style the AI has learned, it's fair to assume it will be realistic and coherent. Although maaaaaaaybe you could detect poisoned images with some dark magic tho, get the targeted AI then analyze the latent space to see if the image has been tampered with

On the other hand, I think if you build more robust features and just scale the data this problems might go away with more regularization in the network. Plus, it assumes you have the target of one AI generation tool, there are a dozen of these, and if someone trains with a few more images in a cluster, that's it, you shifted the features and the poisoned images are invalid

games

Comment on

Terraria developer bashes Unity, donates $200k to open source alternatives

This reminds me of a time when the blender fund was opened and at some point a bunch of companies jumped to donate money (steam, epic, google, AMD...) this was way back when 2.8 was getting in shape. Years later we saw the fruits of that labor with the 3.x series bringing nice improvements and refactors that were done over the course of many months and years

We probably won't see a huge push in godot's quality in what's left of this year, but maybe in 2024 and later

Comment on

YSK : Dark patterns among large companies are becoming more mainstream

Reply in thread

I short of have a theory with this. There's this belief that "netflix killed piracy" because they provided an actual service with a fair price and the commodity that people wanted to watch shows. And that later on, it got enshittified. But I kinda think that, collaterally, a very important factor that explains people not even knowing how to download a torrent or having 0 critical mind when it comes to the other companies abusing their power has been the surge of smartphones

They were designed to have idiot-proof protection, but more and more they distanced newer generations from having a minimal technical background on how to use computers, which then leads to a more ignorant society incapable of saying no to such companies

I'm not saying this has been the main factor but I have my suspicions to believe it might be related