Researchers Told Four AI Models to Hack Nine Others. They Succeeded 97% of the Time.

dev.to

Researchers Told Four AI Models to Hack Nine Others. They Succeeded 97% of the Time.

dev.to

mothasa to

Technology@lemmy.worldEnglish · 2 days ago

Jailbreaking an AI model used to require expertise. You needed to understand prompt engineering,...

Peer-reviewed in Nature Communications: reasoning models autonomously jailbreak other AI systems with 97.14% success rate. No human expertise required. Just a system prompt.

You must log in or register to comment.

Chat

Technology@lemmy.world

technology@lemmy.world

Create a post

You are not logged in. However you can subscribe from another Fediverse account, for example Lemmy or Mastodon. To do this, paste the following into the search field of your instance: [email protected]

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

Visibility: Public

This community can be federated to other instances and be posted/commented in by their users.

1 user / day
24 users / week
24 users / month
54 users / 6 months
0 local subscribers
81.7K subscribers
167 Posts
0 Comments
Modlog