ToolHazard: Scaling Adversarial Environments for Security Evaluation and Alignment of LLM-Based Agents

Large language model LLM agents integrated with external tools are vulnerable to indirect prompt injections embedded in environmental states. However, existing studies largely rely on manually implemented or reused environments, stochastic LLM-based tool simulation, and predefined injection...

We Are Still Unable to Secure LLMs from Malicious Inputs

Nice indirect prompt injection attack: Bargury's attack starts with a poisoned document, which is shared to a potential victim's Google Drive. Bargury says a victim could have also uploaded a compromised file to their own account. It looks like an official document on company meeting policies. Bu...