

Or maybe it’s the other way around 🙀
You may be able to find me on other platforms by the same name!
Mastodon: [email protected]
Contact me on SimpleX or Signal!


Or maybe it’s the other way around 🙀


Language models don’t have any “mind”. No matter how far this goes, current methods (making more and more advanced LLMs) can’t lead to a “superintelligence” or even any kind of real reasoning.


Here’s hoping GrapheneOS decides to support it! 🤞
I was trying to read this while driving, but there wasn’t enough contrast between the white text and the light gray cat 😿
If the people knew how to be invisible, then he wouldn’t be able to see them here


What specifically do you think is hard to avoid? I’ve never accidentally triggered a quick answer, personally


Paying for Yandex APIs is a product of their goal to be the best search engine available. They pay for access to nearly every major search provider and wouldn’t want to lose access to Yandex results just because of the country they’re located in.


Personally, it’s the only search engine I haven’t wanted to get away from
I just hope there will eventually be a solution that doesn’t require tired server owners to sift through registrations and posts all day.
I think the majority of people on Lemmy don’t want to be served an AI-generated feed about AI. Maybe try a more AI- or LLM-focused community?


This really should be marked as NSFW!


What confuses me is what they mean by “corporate VPN data containing unencrypted login details.” Unless the VPN server connects to the backend servers with unencrypted traffic through these satellites (which definitely should not happen) then this should not be possible.


You sould be able to get to it under settings > advanced > “Open Redirects”


To create a redirect:
^https://original.domain.tld/|https://new.domain.tld/
For example, to replace www.youtube.com with inv.nadeko.net (an Invidious instance), you can use the rule
^https://www.youtube.com/|https://inv.nadeko.net/
I’ve also created a copy of this rule for any youtu.be results, as well as X/Twitter and Reddit for their respective frontends. This will work for any pair of websites if you can simply replace the domain in the URL and have it pull up the correct content on the alternative website.


It comes down to the fact that Kagi’s first priority in search is providing the best results, so they’re not going to turn down access to another search api based on politics.


That is also something you can do with Kagi. 😆 I have it set up so that all YouTube results redirect me to Invidious, as well as Reddit and Twitter for their respective frontends.


With some manual configuration, that’s something you can do with Kagi!
Creepypasta and copypasta refer to completely different things
Except current methods can never lead to a “superintelligence.”
I don’t understand, one moment everything’s normal, and then the next day I come back and it’s corn everywhere 😭