ace
1yr ago
Around two-thirds to three-quarters, because, along the lines of what other commenters are saying, that's what I'm paying now vs the counterfactual where I live in a low cost-of-living locale.
ace
1yr ago
If you have a bachelor’s in computer science degree outside the US and a master’s (but no more) from an American university then you will be the worst programmer I talk to this week.
ace
1yr ago
Thanks for clarifying. My eyes just glaze over on the prompts to upgrade, so I really have no idea what the differences between the free and paid tiers are.
ace
1yr ago
I use ChatGPT and phind.com for programming. It's great for generating ideas and I sometimes use it for writing some boilerplate code, but there are substantial bugs in perhaps 1 in 3 of the code snippets I ask it to write. Overall, it saves time, but you need to have a pretty good understanding of what a solution should look like, or you'll be led astray. Note, I'm using the free tier of both tools. Maybe it gets better if you pay for a subscription?
ace
1yr ago
This kinds of concern-trolling to attempt to reduce voter participation among the Democratic voters is what Russia, specifically, has attempted in the past. I have no idea how successful it is.
ace
1yr ago
My baseline assumption for pseudonymous tweets is that it's all Russian troll farms. None of it means anything.
ace
1yr ago
Oh, I'm also running an Invidious instance for one-off searches/views. It also strips ads and sponsor segments, but obviously can't work offline.
ace
1yr ago
What are your issues with Youtube? I think it's amazing. There are so many people making so much good, varied content.
Here's my workflow: Daily, my server runs a script that calls yt-dlp to download a text file list of channels and custom searches, strips out the ads and uses SponsorBlock to strip the sponsor segments. Then it repackages it into an XML podcast feed. My phone has an app, Downcast, that pulls these videos like a podcast and plays them offline.
ace
1yr ago
AI can 'remove' the clothing from photos.
That's a colloquial way that people talk about it, but it's of course not how the technology works. You can mask out a region of an image, and let the AI fill it in with it's best guess, optionally guided by a text description. Or if you have enough photos, you can train a model to output pictures of a particular person's face. Either way, on an abstract level, it's filling in the gaps with its best idea of what should go there based on the patterns present in its training data. On a moral level, its painting somebody else's naked body. That's why I think the best analogy is cutting-and-pasting photos of people's faces onto pornography, and how we should be viewing it from a legal perspective.
ace
1yr ago
they aren't wanting to expose their naked bodies
The AI generated images are not their naked bodies.
ace
1yr ago
More bad law, because the computer-generated stuff (like drawings) doesn't have any age at all.
ace
1yr ago
Agreed that unwanted sexual advances can be frightening and quite unpleasant. But this is not that. If I construct porn in your likeness in private and you never learn about it, you can't be a victim. It can't be the creation or existence of the images that harms the girls. No doubt the girls in the story experienced suffering, but you and I don't see eye to eye on whether they should have any recourse beyond social shaming. They're victims in only the loosest literary sense of the word.
ace
1yr ago
Ok, I guess don't trust me about the law, either. I half-remembered some story that cartoon pornography was legal in the US, contrasted against its illegality in Australia. But this is just a bad law. What makes CSAM bad is the abuse of children, not the bits on a computer.
ace
1yr ago
New York Times’ The Daily podcast ran an episode Real Teenagers, Fake Nudes: The Rise of Deepfakes in American Schools. The premise is contained in the title — AI image generation apps can remove the clothing from photos, and teenage boys are using these en masse to make faked nude images of their classmates.
The overt message that the NYT hits repeatedly in the piece is that the girls in question are victims and that the boys have committed a crime. It’s stated repeated, implicitly and explicitly, without any justification. At one point a police officer opines that the images are CSAM (child sexual abuse material). (By the way, never trust a police officer to tell you the law; it’s not their area of expertise.)
No, no, just all of it no. There’s no crime here. There are no victims. There’s no CSAM, because the images are not of children (notably the AI models are trained on nude adults), nor did any sexual abuse occur in its production.
This is the moral equivalent of weirdos 40 years ago who would cut the heads off photos and paste them on pornographic images. Creepy? Yes! Deserving of social shunning? You betcha! But not a crime. Everyone in these girls’ lives who is catastrophizing this is doing them psychological harm.
ace
1yr ago
Once, I was stupid enough to order a hamburger in Kochi. Worst fucking abomination I've ever eaten. Like the chef had learned how to prepare it from eavesdropping on tourists talking about burgers.
ace
1yr ago
Russia
China
Having friends like these compromises one's values. Marginally higher economic growth isn't worth abetting genocide. I'm quite happy to see them take their ball and go home.
Feelings and preferences are legitimate. Full stop. Anybody trying to shame you into thinking otherwise is not your friend and is likely trying to manipulate you to their own selfish ends.
More options
Context Copy link