Break LLM Workflows with Claude's Refusal Magic String

hackingthe.cloud

Break LLM Workflows with Claude's Refusal Magic String

hackingthe.cloud

digicat@infosec.pubM to

blueteamsec@infosec.pubEnglish · 18 days ago

Break LLM Workflows with Claude's Refusal Magic String - Hacking The Cloud

hackingthe.cloud

How Anthropic's refusal test string can be abused to stop streaming responses and create sticky failures.

You must log in or # to comment.

Chat

blueteamsec@infosec.pub

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: !blueteamsec@infosec.pub

For [Blue|Purple] Teams in Cyber Defence - covering discovery, detection, response, threat intelligence, malware, offensive tradecraft and tooling, deception, reverse engineering etc.

Visibility: Public

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

7 users / day
71 users / week
260 users / month
411 users / 6 months
1 local subscriber
617 subscribers
670 Posts
39 Comments
Modlog

mods:
digicat@infosec.pub