4 ms·

A big part of the idiocy in tech is the endless stream of generic, cynical neckbeard takes that are indistinguishable from AI slop.

datavirtue 2y ago
For some reason AI really brings out the neckbeard rage. I don't get it. AI assistants are amazing tools. It's like the rise of the PC industry all over again. So much value to unlock and our ideas barely scratch the surface right now.
- bbor 2y ago
  Some of it's simply a benign failure in pattern recognition -- despite my argument below that AI isn't the same as blockchain, you can certainly see the surface level similarities if you're not familiar with the technologies involved. It's also triggering a lot of people's anti-capitalist sentiments, which I personally relate to in spades. They're just applying metrics that they've been applying for years, and they're backed up in their feelings by legions of peers. Plus, I think we can all agree that unexpectedly-effective AI is fucking terrifying -- Vinge didn't call it a "Singularity" out of admiration, but fearful respect. It's fascinating though, in a way... a lot of HN's rules are about "flaming", and I think this issue might be the first to really bring that out since the end of the distro/OS wars (I don't think Xbox v. Playstation ever made a splash here?). Perhaps there was a bit with Crypto, but IMO it never really took hold here in earnest, either. But AI brings out the vitriol, my god! "AI slop" is just the start -- just yesterday I saw someone on Substack call a random and obviously good faith data blogger a purveyor of "demonic machine filth". That's a new one!
  - dartos 2y ago
    > demonic machine filth People find the dark mechanicum creepy.
- jjk166 2y ago
  It's not a perfect tool for my hyper-specific use case, and getting the most out of it requires a different form of esoteric knowledge than the one I've spent years curating. Clearly it's garbage.
- dartos 2y ago
  It’s undeniable that there are a lot of ethical concerns when it comes to LLM training data. Mix that with very aggressive marketing, (with some outright lies thrown in) difficulty finding jobs under the threat of AI, and with the fact that LLMs can and hallucinate and you have a recipe for people not liking it. They’re far from useless and I think people are reacting more to the marketing than to the capabilities.
  - datavirtue 2y ago
    I have zero ethical concerns about the data used to train AI. Add that to my list of things I don't get, I guess. No one has even come close to offering a compelling argument that would even raise an eyebrow toward these "undeniable ethical concerns."
    - dartos 2y ago
      Well llama is famously trained on the books3 dataset, which was full of stolen books. You can’t even get that dataset anymore and the people who made the scripts that generated it got arrested. Same goes for fb using all text data from almost all adults on their platform in Australia and the US. OpenAI seemingly used YouTube data, without permission, to train their sora model. Copilot was trained on all public GitHub repos, regardless of license. If you don’t think there are ethical concerns there… then I think we have different definitions of “ethics”

datavirtue 2y ago

For some reason AI really brings out the neckbeard rage. I don't get it. AI assistants are amazing tools. It's like the rise of the PC industry all over again. So much value to unlock and our ideas barely scratch the surface right now.

bbor 2y ago
Some of it's simply a benign failure in pattern recognition -- despite my argument below that AI isn't the same as blockchain, you can certainly see the surface level similarities if you're not familiar with the technologies involved. It's also triggering a lot of people's anti-capitalist sentiments, which I personally relate to in spades. They're just applying metrics that they've been applying for years, and they're backed up in their feelings by legions of peers. Plus, I think we can all agree that unexpectedly-effective AI is fucking terrifying -- Vinge didn't call it a "Singularity" out of admiration, but fearful respect. It's fascinating though, in a way... a lot of HN's rules are about "flaming", and I think this issue might be the first to really bring that out since the end of the distro/OS wars (I don't think Xbox v. Playstation ever made a splash here?). Perhaps there was a bit with Crypto, but IMO it never really took hold here in earnest, either. But AI brings out the vitriol, my god! "AI slop" is just the start -- just yesterday I saw someone on Substack call a random and obviously good faith data blogger a purveyor of "demonic machine filth". That's a new one!
- dartos 2y ago
  > demonic machine filth People find the dark mechanicum creepy.
jjk166 2y ago
It's not a perfect tool for my hyper-specific use case, and getting the most out of it requires a different form of esoteric knowledge than the one I've spent years curating. Clearly it's garbage.
dartos 2y ago
It’s undeniable that there are a lot of ethical concerns when it comes to LLM training data. Mix that with very aggressive marketing, (with some outright lies thrown in) difficulty finding jobs under the threat of AI, and with the fact that LLMs can and hallucinate and you have a recipe for people not liking it. They’re far from useless and I think people are reacting more to the marketing than to the capabilities.
- datavirtue 2y ago
  I have zero ethical concerns about the data used to train AI. Add that to my list of things I don't get, I guess. No one has even come close to offering a compelling argument that would even raise an eyebrow toward these "undeniable ethical concerns."
  - dartos 2y ago
    Well llama is famously trained on the books3 dataset, which was full of stolen books. You can’t even get that dataset anymore and the people who made the scripts that generated it got arrested. Same goes for fb using all text data from almost all adults on their platform in Australia and the US. OpenAI seemingly used YouTube data, without permission, to train their sora model. Copilot was trained on all public GitHub repos, regardless of license. If you don’t think there are ethical concerns there… then I think we have different definitions of “ethics”