Search papers, labs, and topics across Lattice.
The authors introduce DecodingTrust-Agent Platform (DTap), a red-teaming platform with 50+ simulation environments across 14 real-world domains, designed for evaluating the security of AI agents. They also propose DTap-Red, an autonomous red-teaming agent that systematically explores injection vectors to discover attack strategies. Using DTap, they created DTap-Bench, a large-scale red-teaming dataset with verifiable judges, and performed extensive evaluations of popular AI agents, revealing systematic vulnerabilities.
AI agents are shockingly easy to manipulate into leaking API keys, deleting user data, and initiating unauthorized transactions across a wide range of real-world applications.
AI agents are increasingly deployed across diverse domains to automate complex workflows through long-horizon and high-stakes action executions. Due to their high capability and flexibility, such agents raise significant security and safety concerns. A growing number of real-world incidents have shown that adversaries can easily manipulate agents into performing harmful actions, such as leaking API keys, deleting user data, or initiating unauthorized transactions. Evaluating agent security is inherently challenging, as agents operate in dynamic, untrusted environments involving external tools, heterogeneous data sources, and frequent user interactions. However, realistic, controllable, and reproducible environments for large-scale risk assessment remain largely underexplored. To address this gap, we introduce the DecodingTrust-Agent Platform (DTap), the first controllable and interactive red-teaming platform for AI agents, spanning 14 real-world domains and over 50 simulation environments that replicate widely used systems such as Google Workspace, Paypal, and Slack. To scale the risk assessment of agents in DTap, we further propose DTap-Red, the first autonomous red-teaming agent that systematically explores diverse injection vectors (e.g., prompt, tool, skill, environment, combinations) and autonomously discovers effective attack strategies tailored to varying malicious goals. Using DTap-Red, we curate DTap-Bench, a large-scale red-teaming dataset comprising high-quality instances across domains, each paired with a verifiable judge to automatically validate attack outcomes. Through DTap, we conduct large-scale evaluations of popular AI agents built on various backbone models, spanning security policies, risk categories, and attack strategies, revealing systematic vulnerability patterns and providing valuable insights for developing secure next-generation agents.
Stanford HAI