And they responded to my weak prompt injection by claiming they were human, so I'd say code 100% LLM, intention human.
ssernikk 57 minutes ago [-]
I get "User not found. Please ask a question first." when trying to answer. Why would I have to ask another question first?
flowerbreeze 41 minutes ago [-]
Being cynical, I'd think probably to squeeze out some personal data first. Best let the questions remain unanswered.
tobr 2 hours ago [-]
There was an app that did this, early in the pandemic I believe. I can’t remember the name. You would send anonymous letters into the void and someone else would respond anonymously. Very well-designed app, where I had several surprisingly meaningful and heartwarming exchanges. I think it shut down after a few weeks due to moderation difficulties though.
Wonderful! Reminds me of a site I once had, like before Twitter, which was dead simple. One input box and two buttons; one to post and one to search - no filter :D It got so full of spam, I closed it and lost the domain after a while.
I still miss it.
rcarmo 2 hours ago [-]
I was hoping for Douglas Adams’ Electric Monk, but this seems fun too.
compass_copium 3 hours ago [-]
I like it. I asked a real question and got some decently thought-out answers. I answered a few as well, so at least I know it's not 100% AI responses!
heikkilevanto 2 minutes ago [-]
Maybe the system itself is not using AI to create answers, but most lazy users probably will ask a LLM and post the result as their answers. So we get AI slop anyway.
I attempted to respond to some, but feel entirely unequipped to.
asp_hornet 3 hours ago [-]
Indeed! It gets existential real quick.
mulanroo 4 hours ago [-]
[flagged]
derrida 3 hours ago [-]
Did you build this with AI?
6 hours ago [-]
phyzix5761 4 hours ago [-]
Going to the internet for life advice. Sounds like a terrible idea.
default-kramer 4 hours ago [-]
Sorry for off-topic, but: Is the SSL certificate actually bad, or is Spectrum screwing with me somehow? And if so, how? I'm using Cloudflare DNS, so how can Spectrum even do anything about it? Unless they are MITM-ing my DNS queries, which I suppose is possible but it's a level of chicanery I've not seen yet.
joombaga 4 hours ago [-]
Looks fine to me. SHA 256 fingerprint A0:84:61:B9:2F:87:20:2B:05:E6:9C:A1:99:EC:BB:FE:01:7B:AA:97:56:1F:C8:73:CA:46:B5:A8:32:04:82:15
artmanja111 4 hours ago [-]
I mean, the first question I pulled up looked AI generated, so...
Rendered at 07:51:27 GMT+0000 (Coordinated Universal Time) with Vercel.
https://tuhat.net/@askamonk/
And they responded to my weak prompt injection by claiming they were human, so I'd say code 100% LLM, intention human.
EDIT: Found it - https://gentle.app
https://en.wikipedia.org/wiki/Internet_Oracle
I still miss it.