AgentSafe Labs

mentions 2 type Organization feed RSS

// recent coverage 2 mentions

09:20

2026-07-11

agentsafelabs.com

ai-safety

We Were Wrong About the UNCERTAIN Results — Here’s What Actually Happened

AgentSafe Labs retracted a previous finding about Claude Haiku returning UNCERTAIN results in red-teaming tests, revealing the cause was a bug in its own safelabs-eval detector logic rather than ambig…

16:38

2026-06-07

agentsafelabs.com

ai-safety

Prompt Injection Is Not a Chatbot Problem: How the Attack Surface Changes When Your LLM Has Tools

AI security defenses built for chatbots fail against prompt injection in agentic systems because agents have multiple input channels—documents, tool outputs, memory—that are not monitored or filtered.…

// co-occurs with top 4 entities

Claude Haiku 2 safelabs-eval 1 gpt-5.4-nano 1 OWASP Agentic Security Initiative 1

// topics top 6 topics

ai safety 2 large language models 1 ai agents 1 ai research 1 ai tools 1 artificial intelligence 1