Isn’t there code / the magic incantation of prompt text that we can deploy to get bots to reveal themselves? Even if it take more than one response?
this post was submitted on 05 Sep 2024
565 points (96.2% liked)
Technology
60450 readers
5326 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related content.
- Be excellent to each another!
- 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, to ask if your bot can be added please contact us.
- Check for duplicates before posting, duplicates may be removed
Approved Bots
founded 2 years ago
MODERATORS
To help fight bot disinformation, I think there needs to be an international treaty that requires all AI models/bots to disclose themselves as AI when prompted using a set keyphrase in every language, and that API access to the model be contingent on paying regain tests of the phrase (to keep bad actors from simply filtering out that phrase in their requests to the API).
It wouldn't stop the nation-state level bad actors, but it would help prevent people without access to their own private LLMs from being able to use them as effectively for disinformation.
load more comments
(2 replies)