DeepTeam is a framework to red team LLMs and AI agents.
updated blog
A-Vamshi6eb8fd3f2d2d
diff --git a/docs/blog/intensive-alignment-paradox.md b/docs/blog/intensive-alignment-paradox.mdindex 16cdc10e..3541c613 100644--- a/docs/blog/intensive-alignment-paradox.md+++ b/docs/blog/intensive-alignment-paradox.md@@ -2,7 +2,7 @@title: Intensive Alignment Training Backfiresdescription: Systematic analysis revealing how intensive safety training in modern reasoning models may create exploitable vulnerabilities, making heavily-aligned models more susceptible to sophisticated attacks than their predecessorsslug: ai-safety-paradox-deepteam-authors: [sid]+authors: [sid, penguine]date: 2025-06-01hide_table_of_contents: false# TODO: Add proper image path once available