• 0 Posts
  • 28 Comments
Joined 1 year ago
cake
Cake day: June 16th, 2023

help-circle




  • Here’s what would happen in capitalist America: entities would own those machines and use them as a means of personal enrichment, it’d displace a ton of human workers, the taxes generated from profits generated wouldn’t offset the economic impacts, and then half of the lawmakers would introduce bills that would provide lucrative incentives to those entities if they maintain a certain ratio of human workers and they’d staple a bunch of regressive crap onto it like abortion or whatever, it wouldn’t pass because the other half of lawmakers would want to tax the hell out of profits made with those machines, government would shut down 4 times a year, Jeff Bezos builds a vacation home on the moon











  • Quora did something similar to that. On mobile it’ll pester you to download the app, and on the app it’ll pester you to download an AI app, in an attempt to keep people bouncing around between those 2 apps.

    Quora seems to suffer from the same issue as Reddit. AI can take a single page view and distribute the information obtained from that single page view multiple times, basically cut out the source in terms of monetization, and those 2 sites benefit the most from people using Google to find something. I think Quora’s approach is a bit more sensible than Reddit’s, basically jumping on the AI bandwagon as opposed to cutting off both AI and people using Google


  • I think in Reddit’s case, mostly due to the garbage search function, they’re stuck deciding between shooting themselves in the foot right now by blocking Google, or allowing generative AI to shoot them in the foot later by allowing Google to scrape user content. There will soon come a day when less people use Google to find specific information on Reddit and more people ask an AI prompt and get the same information fed to them without having to click through to Reddit.

    I think the most ideal way to coexist with AI is for Reddit to pull out of search engines and make their own search engine work well, but until then they’re in a pickle




  • I’m not gonna google the whole ‘GDP’ franchise/entity for hopefully obvious reasons but the ‘GDP’ franchise very well could be a verified account that creates content through exploitation and that being a very real possibility punches a hole in the effectiveness of attempting to moderate content by only allowing content derived from verified accounts. The article notes that the actual people reviewing flagged content can’t handle the sheer amount of flagged content so it’s likely they can’t handle verifying that content created by verified accounts isn’t content created through exploitation, so it’s probable once an account is verified it can just start sliding in all sorts of stuff that will go unchecked (as it apparently did in this case and others).