Technology

69772 readers

3612 users here now

This is a most excellent place for technology news and articles.

Our Rules

Follow the lemmy.world rules.
Only tech related news or articles.
Be excellent to each other!
Mod approved content bots can post up to 10 articles per day.
Threads asking for personal tech support may be deleted.
Politics threads may be removed.
No memes allowed as posts, OK to post as comments.
Only approved bots from the list below, this includes using AI responses and summaries. To ask if your bot can be added please contact a mod.
Check for duplicates before posting, duplicates may be removed
Accounts 7 days and younger will have their posts automatically removed.

Approved Bots

founded 2 years ago

MODERATORS

[email protected]

1012

Somebody managed to coax the Gab AI chatbot to reveal its prompt (infosec.exchange)

submitted 1 year ago by [email protected] to c/[email protected]

269 comments fedilink hide all child comments

(page 4) 50 comments

sorted by: hot top controversial new old

[–] [email protected] 6 points 1 year ago

I managed to get partial prompts out of it then... I think It's broken now:

[–] [email protected] 6 points 1 year ago (2 children)

I don't get it, what makes the output trustworthy? If it seems real, it's probably real? If it keeps hallucinating something, it must have some truth to it? Seems like the two main mindsets; you can tell by the way it is, and look it keeps saying this.

load more comments (2 replies)

[–] [email protected] 4 points 1 year ago (3 children)

This seems like a lot of detail... like maybe too much detail for it to be real??

load more comments (3 replies)

[–] [email protected] 4 points 1 year ago (4 children)

Looks like they caught on. It no longer spews its prompt. At least, not for me.

[–] [email protected] 4 points 1 year ago

Just worked for me, I think you just got unlucky

load more comments (3 replies)

[–] [email protected] 4 points 1 year ago (2 children)

Tried to use it a bit more but it's too smart...

load more comments (2 replies)

[–] [email protected] 4 points 1 year ago (1 children)

I think it is good to to make an unbiased raw "AI"

But unfortunately they didn't manage that. At least is some ways it's a balance to the other AI's

[–] [email protected] 5 points 1 year ago* (last edited 1 year ago) (2 children)

I think it is good to to make an unbiased raw "AI"

Isn't that what MS tried with Tai and it pretty quickly turned into a Nazi?

[–] [email protected] 6 points 1 year ago (1 children)

Tai was actively being manipulated by malicious users.

[–] [email protected] 5 points 1 year ago* (last edited 1 year ago)

That's fair. I just think it's funny that the well intentioned one turned into a Nazi and the Nazi one needs to be pretty heavy handedly told not to turn into a decent "person".

[–] [email protected] 5 points 1 year ago (1 children)

Tay tweets was a legend.

That worked differently though they tried to get her to learn from users. I don't think even chat GPT works like that.

load more comments (1 replies)

load more comments