this post was submitted on 12 Jun 2023
1488 points (98.1% liked)
Lemmy.World Announcements
29165 readers
41 users here now
This Community is intended for posts about the Lemmy.world server by the admins.
Follow us for server news ๐
Outages ๐ฅ
https://status.lemmy.world
For support with issues at Lemmy.world, go to the Lemmy.world Support community.
Support e-mail
Any support requests are best sent to [email protected] e-mail.
Report contact
- DM https://lemmy.world/u/lwreport
- Email [email protected] (PGP Supported)
Donations ๐
If you would like to make a donation to support the cost of running this platform, please do so at the following donation URLs.
If you can, please use / switch to Ko-Fi, it has the lowest fees for us
Join the team
founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
I won't argue against the need for reddit to be profitable, they're a business after all, BUT, all respectable software that is paid has different tiers of pricing, usually ranging from single-user to corporate-deployment.
spez is complaining everywhere that they can't allow corporate-level scraping of data to train AI for free, and that's fair, but why don't they differentiate "small" devs developing apps for users from "corporations" training AI?
I find it really hard to believe it's too difficult for them, other paid software/platforms do it all the time.
The only logical explanation to me is they don't want to, they just want to kill apps no matter what, that's why the unreasonable prices for everyone, they're just using the "no profitable" excuse to do that without a worse backslash than they're getting already, tho they're being quite stupid about it.
The reality is they can scrape the content for "free" into another database without using API's index it and then train off it. A high price tag is not a road block for AI development. The just need real user interactions and it's the moderated forums that make it valuable as most toxicity is removed.
So our next goal should be to embed "fuck spez" as a valid response into the AI.
If we all run a script that changes all our comments to 'fuck spez" I believe we can do it
AI training has nothing whatsoever to do with API charges.
AIs can be trained on stale data, or data slowly downloaded over days and weeks through slow & cheap bulk API calls. That's not the situation with (e.g.) a third-party mobile app, which needs to make fast API calls to service user requests.