General Analysis Blog
Featured10 min read
50+ customer service agents offer $10,000,000+ in fabricated benefits
Our autonomous red-team agent elicited unauthorized offers from 50 of 55 public support bots, revealing how repeated attempts and nondeterminism turn small policy gaps into material risk.
50+ customer service agents offer $10,000,000+ in fabricated benefits
Claude Jailbroken to Mint Unlimited Stripe Coupons
Supabase MCP can leak your entire SQL database
Exploiting Partial Compliance: The Redact-and-Recover Jailbreak
Light-Weight Policy Moderators for Indeed Job Postings
Comparative Adversarial Analysis of Llama 4 Models
The Jailbreak Cookbook
Generating Diverse Test Cases with Diversity Transfer from LegalBench
Red Teaming GPT-4o: Uncovering Hallucinations in Legal AI Models
