Why do people use grok?
Discussion
Because they hang around on X anyway all day?
I guess so. I wonder if Nostr has an LLM implementation? Does it even really need one?
Maybe a replybot would be kind of great as you call the LLM with a question and tag or @ and then it replies with the content.
nostr:npub1de6l09erjl9r990q7n9ql0rwh8x8n059ht7a267n0q3qe28wua8q20q0sd what do you think?
You want to build on your LLM models and hook it up?
We had integrated LLMs before X. Nakamoto and Cyber(whatshisname?). Just nobody using them any more by the looks of it. As for LLM in general, the new thing is personalised SLM connected to LLMs. Train your own assistant.
I didn't even know; damn. I guess a lot of the stuff is getting lost in the weeds out here. Do you know how to call the LLM? was it an @ or a # ? I wonder if they are still active
nostr:npub13wfgha67mdxall3gqp2hlln7tc4s03w4zqhe05v4t7fptpvnsgqs0z4fun used to drop walls of text in the zapstream chat every time someone used the words funny and joke. Aint that right Cyborg?
Also, can't find the npub for NakamotoX. He was better than Cyborg
The same reason they use any other AI, to get informationβ¦
and yet Grok has notoriously been cooked to lie and spread misinformation?
most of these people who use grok haven't the slightest clue what is happening, they don't know they are being fed misinformation⦠the others are Elon's butt lickers
not saying all LLMs don't carry bias or misinformation datasources, but Grok is an explicit case of egregious use where it, not only, openly admits to it, but has inserted the misinfomation content into unrelated replies making it stand out even more.
Yet, I still see it here on Nostr, and more importantly, people are still okay with using it.
For its integration in X
so if Nostr had an LLM, people would opt to use that instead?
Maybe. But the main usecase for Grok is access to X posts which others don't have. People often ask Grok to explain posts, give context, fact check etc..
I've personally found Grok shit
The algo rules all. Monkey see (a lot); monkey do.
I personally donβt like Sam Altman/Open AIβs practices. Better the devil/developer you know. Iβd be using Mistral, but conversation mode is a must.
100%. It's subjective really, but that's why I wanted to know.
I feel that Grok is the most egregious of the pack, but there's open source LLMs and community ones too which are much better.