2026-05-14 09:34 UTC
@pelle@veganism.social @xjasminelu@hci.social @radicle@toot.radicle.dev
side note:
I work under the assumption that technology cannot be taken back easily
once the cat's out of the bag, that's it
imo llms fundamentally suck by being spammy
framing them tight can alleviate some issues and rigorous benchmarks can be used to objectively measure results
this is what's happening at community-computer:
https://radicle.network/nodes/iris.radicle.xyz/rad%3Az4Wk8hdpwG4HtoCxr1uuoQDpnfr25
even if all the datacenters experienced rapid unscheduled disassembly this would be a method that generated traceable computing experiments
and even failed experiments - if recorded methodically - always render valuable data
Replies (2)
-
@xjasminelu@hci.social 2026-05-14 09:52
@roy_calum@universeodon.com @pelle@veganism.social @radicle@toot.radicle.dev thank you for the detailed write up! I’m a huge Luddite when it comes to AI (I can be /many folks unfortunately gotta spend the tokens to pay the bills) so am especially interested in all the alternative “experiments” in this new world we’re in w AI
-
@pelle@veganism.social 2026-05-14 09:53
@roy_calum@universeodon.com > framing them tight can alleviate some issues and rigorous benchmarks can be used to objectively measure results please don't sell me #AIslop 💩✋